Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Development and characterization of methotrexate loaded poly(L-lactic acid) microspheres for the treatment… Liang, Sanching Linda 2006

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2006-130164.pdf [ 16.72MB ]
Metadata
JSON: 831-1.0092780.json
JSON-LD: 831-1.0092780-ld.json
RDF/XML (Pretty): 831-1.0092780-rdf.xml
RDF/JSON: 831-1.0092780-rdf.json
Turtle: 831-1.0092780-turtle.txt
N-Triples: 831-1.0092780-rdf-ntriples.txt
Original Record: 831-1.0092780-source.json
Full Text
831-1.0092780-fulltext.txt
Citation
831-1.0092780.ris

Full Text

DEVELOPMENT AND CHARACTERIZATION OF M E T H O T R E X A T E LOADED POLYfL-LACTIC ACID) MICROSPHERES FOR T H E TREATMENT OF RHEUMATOID ARTHRITIS by Sanching Linda Liang B.Sc. (Pharm), University of British Columbia, 1999  A THESIS SUBMITTED IN P A R T I A L F U L F I L M E N T OF T H E REQUIREMENTS FOR THE D E G R E E OF DOCTOR OF PHILOSOPHY in  T H E F A C U L T Y OF G R A D U A T E STUDIES  (Pharmaceutical Sciences)  T H E UNIVERSITY OF BRITISH C O L U M B I A November 2005 © Sanching Linda Liang, 2005  ABSTRACT Methotrexate (MTX) has shown anti-inflammatory effects in the treatment of rheumatoid arthritis.  Attempts by other groups have been made to improve the efficacy  and reduce toxicity by administering the drug intra-articularly, but the outcomes were not successful due to rapid clearance of the drug from the joint cavity. M T X loaded polymeric microspheres may provide a controlled release drug delivery system to maintain an effective concentration of M T X in the joint cavity. The goals were to develop M T X loaded microspheres and to determine the in vivo biodistribution and efficacy following intra-articular injection in rabbit joints.  M T X loaded poly(L-lactic  acid) microspheres (size range 33-110 pm) manufactured from poly(L-lactic acid) with an average molecular weight of 2000 g/mole (PLLA2k) showed good tolerability in rabbit joints.  The in vitro drug release profiles of M T X loaded P L L A 2 k microspheres  demonstrated a rapid burst phase with more than 50% of drug being released within 5 days followed by a slow release phase. Pharmacokinetics of M T X following intra-articular injection of both 1.5 mg and 10 mg doses of either M T X solution or M T X loaded microspheres (33-110 pm) were investigated in healthy rabbits.  Plasma concentration peaked at 15 min (t  max  ) following  intra-articular injection, and the maximum plasma concentration (C ) for rabbits max  injected with M T X solution was 5 fold higher than for rabbits injected with M T X ii  microspheres.  Approximately 70% of injected M T X dose was excreted in the urine of  the rabbits injected with M T X solution while only 12% of the dose was excreted in the urine of the rabbits injected with M T X microspheres 24 h following intra-articular injection. In vivo efficacy of intra-articular M T X loaded PLLA2k microspheres (33-110 um) or M T X solution was evaluated using an antigen-induced arthritis rabbit model. Arthritis was successfully induced in the joints of rabbits with the observation of histopathological features resembling rheumatoid arthritis.  Based on the degree of  swelling of the knee joints and a system of scoring the pathological features of the disease, there was no significant difference between M T X solution and microspheres treated groups compared to phosphate buffered saline (control) animals.  The lack of  therapeutic responses to M T X loaded microspheres treatment was likely due to the severity of the disease induced and insufficient length of the observation period. M T X loaded P L L A 2 k microspheres were shown to be well tolerated in the rabbit knee joints and provide a controlled, localized delivery of M T X into the joint cavity following intra-articular injection.  iii  T A B L E OF CONTENTS  ABSTRACT  ii  T A B L E OF CONTENTS  iv  LIST OF TABLES  ix  LIST OF FIGURES  xii  LIST OF ABBREVIATIONS  xxi  ACKNOWLEDGEMENTS  xxiii  Chapter 1  1  1.1  PROJECT OVERVIEW  1  1.2  SYNOVIAL JOINT STRUCTURE  3  1.2.1  Synovial joint anatomy  3  1.2.2  Synovium  6  1.2.3  Articular cartilage  7  1.2.4  Synovial fluid  8  1.3  RHEUMATOID ARTHRITIS  11  1.3.1  Pathogenesis  11  1.3.2  Clinical presentation  15  1.3.3  Management of the disease  15  1.3.3.1 Nonsteroidal anti-inflammatory drugs  1.4  ".  16  1.3.3.2 Corticosteroids  17  1.3.3.3 Disease modifying anti-rheumatic drugs  17  1.3.3.4 Biologic disease modifying anti-rheumatic drugs  18  METHOTREXATE  1.4.1  21  Chemistry  21 iv  1.4.2  Formulations  24  1.4.3  Pharmacology and indications  24  1.4.4  Toxicity  28  1.4.5  Pharmacokinetics  31  1.5  INTRA-ARTICULAR DRUG DELIVERY  34  1.5.1  Synovial permeability and effects of inflammation  34  1.5.2  Intra-articular drug therapy  35  1.6  POLYMERIC DRUG DELIVERY SYSTEMS  40  1.6.1  Controlled release drug delivery systems  40  1.6.2  Biodegradable polymers for drug delivery  40  1.6.2.1 Structure and molecular weight of polymers  43  1.6.2.2 Polymer morphology  47  1.6.2.3 Thermal transitions  50  1.6.2.4 Biodegradation  52  1.6.2.5 Biocompatibility  54  1.6.3  Mechanisms of drug release from polymeric drug delivery systems  1.6.3.1 Diffusion controlled drug release  57  1.6.3.2 Diffusion and degradation controlled drug release  59  1.6.3.3 Swelling controlled drug release  60  1.6.4 1.7  56  Factors affecting drug release from biodegradable polymers  60  INTRA-ARTICULAR MICROSPHERES AS A DRUG DELIVERY  SYSTEM  63  1.7.1  Microspheres composition and size  63  1.7.2  Pharmacokinetics and efficacy studies  67  1.8  THESIS GOALS AND OBJECTIVES  69  Chapter 2  71  2.1  71  INTRODUCTION  2.2  EXPERIMENTAL  76  2.2.1  Materials  76  2.2.2  General equipment and supplies  76  2.2.3  Validation of UV-vis spectrophotometric assay  78  2.2.4  Solubility of methotrexate  ,  79  2.2.5  Stability of M T X in PBS at 37°C.....:....  79  2.2.6  Preparation of M T X loaded microspheres  80  2.2.7  Encapsulation efficiency  81  2.2.7.1 Validation of encapsulation efficiency studies 2.2.8  Microspheres characterization  82  2.2.8.1 Particle size determination  82  2.2.8.2 X-ray powder diffraction of microspheres  82  2.2.8.3 Scanning electron microscopy  82  2.2.8.4 Thermal properties of microspheres  83  2.2.9  In vitro release of M T X from P L L A microspheres  2.2.10 Degradation study of microspheres 2.2.10.1 Gel permeation chromatography (GPC) 2.2.11 2.3  81  In vivo tolerability of M T X loaded microspheres in rabbit joints  RESULTS  83 84 84 85 87  2.3.1  Validation of UV-vis spectrophotometry assay for M T X in PBS  87  2.3.2  Solubility and chemical stability of M T X  90  2.3.3  Optimization of manufacturing of M T X loaded microspheres  91  2.3.4  In vitro drug release profiles  96  2.3.4.1 The effect ofpolymer molecular weight on MTX release  96  2.3.4.2 The effect of MTX loading on MTX release  97  2.3.5.3 The effect of gamma irradiation on drug release  97  2.3.5  101  2.3.4.2 Surface morphology  101  2.3.4.3 X-ray powder diffraction patterns of PLLA and PLLA microspheres  101  2.3.4.4 Thermal properties  101  2.3.6  Degradation of microspheres  107  2.3.6.1 Thermal analysis of degraded microspheres  112  2.3.6.2 Surface morphology of degraded microspheres  112  2.3.7 2.4  Characterization of microspheres  Biocompatibility of M T X loaded and control microspheres  DISCUSSION  113 115  Chapter 3  128  3.1  128  INTRODUCTION  vi  3.2  EXPERIMENTAL  132  3.2.1  Materials  132  3.2.2  Preparation of M T X loaded microspheres  132  3.2.3  M T X and 7-OH-MTX assays  133  3.2.3.1 MTX and 7-OH-MTX extraction  133  3.2.3.2 HPLC assay for plasma and tissue samples  134  3.2.3.3 HPLC assay for urine samples  135  3.2.3.4 Recovery of MTX and 7-OH-MTX from tissues  136  3.2.3.5 Validation of the assays  136  3.2.4  Microspheres recovery from joint fluids  137  3.2.5  Pharmacokinetic studies in rabbits  138  3.2.5.1 Low dose pharmacokinetic study  138  3.2.5.2 High dose pharmacokinetic study  139  3.2.6  Pharmacokinetic calculations  140  3.2.7  Histological analysis  142  3.2.8  A S T analysis of rabbit plasma samples  143  3.2.9  Statistical analysis  144  3.3  RESULTS  3.3.1  146  M T X and 7-OH-MTX assay validation  146  3.3.1.1 Specificity, stability, and recovery  146  3.3.1.2 Precision, accuracy, linearity, range, and limit of detection  147  3.3.2  Microspheres recovery from rabbit joint  3.3.3  155  Pharmacokinetic studies in rabbits  157  3.3.3.1 Low dose pharmacokinetic study  157  3.3.3.2 High dose pharmacokinetic study  161  3.3.5  Histological analysis  172  3.3.6  A S T analysis  175  3.4  DISCUSSION  177  Chapter 4  188  4.1  INTRODUCTION  188  4.2  EXPERIMENTAL  193  4.2.1  Materials  4.2.2  Animals and housing  193 : vii  194  4.2.3  Animal care and use committee approval  194  4.2.4  Induction of rheumatoid arthritis  194  4.2.5  Intra-articular injection and blood collection from the rabbits  195  4.2.6  Analysis of M T X and 7-OH-MTX in rabbit plasma  197  4.2.7  H P L C analysis of M T X and 7-OH-MTX  197  4.2.8  Pharmacokinetic calculations  198  4.2.9  Statistical analysis  198  4.2.10  Monitoring of arthritis development  198  4.2.11 Histological processing of knee joints  199  4.2.12  202  4.3  Grading histological slides  RESULTS  204  4.3.1  M T X and 7-OH-MTX plasma concentrations  204  4.3.2  Joint swelling evaluation  213  4.3.3  Histological Analysis  216  4.4  DISCUSSION  231  Chapter 5  239  5.1  SUMMARIZING DISCUSSION  239  5.2  CONCLUSIONS  246  5.3  SUGGESTIONS FOR FUTURE WORK  247  REFERENCES  250  viii  LIST OF TABLES Table 1.1  Adverse effects when using methotrexate for the treatment of rheumatoid arthritis (Furst, 1997)  30  Table 2.1  List of general equipment and supplies used in the study  Table 2.2  Intra-day and inter-day precision of UV-vis spectrophotometry assay of M T X in phosphate buffer saline (pH 7.4)  Table 2.3  77  88  Accuracy of UV-vis spectrophotometry assay of M T X in phosphate buffer saline solution (pH 7.4)  89  Table 2.4 Percentage degradation of M T X in phosphate buffered saline (pH. 7.4) following 7 days of storage at 37°C. The values are mean ± standard deviation (n=4) Table 2.5  90  Manufacturing conditions and properties of M T X loaded microspheres manufactured from 2k, 50k and 100k g/mole P L L A polymer  Table 2.6  Percentage yield of microspheres manufactured from PLLA2k, 50k and 100k following sieving between 33 um, and 110 pm sieves  Table 2.7  93  94  Thermal properties of control and 10% M T X loaded P L L A 2 k microspheres following in vitro degradation in 0.1m PBS (pH 7.4) at 3 7°C.  Table 2.8  Microspheres are in the size range of 3 3 -110 pm  106  Joint swelling and histological analysis of proteoglycan loss in cartilage of rabbit joints injected intra-articularly with 25 mg control or M T X loaded P L L A 2 k microspheres in the size range of 33-110 urn. Joint swelling and proteoglycan loss were scored on a 0 to 4 point scoring system  Table 3.1  114  Percentage of M T X recovered from tissues, plasma and urine samples spiked with known concentrations of M T X following extraction and H P L C analysis. The values are shown as the mean of 3 samples ± one standard deviation (n=3)  Table 3.2  149  Inter- and intra-day precision for the H P L C assay with fluorescence detection for A) M T X , B) 7-OH-MTX extracted from rabbit plasma.... 152  Table 3.3  Accuracy for the H P L C assay with fluorescence detection for A) M T X , B) 7-OH-MTX extracted from rabbit plasma  ix  153  Table 3.4  Inter- and Intra-day precision of the H P L C assay with U V detection for A) M T X , B) 7-OH-MTX extracted from rabbit urine  Table 3.5  154  The amounts of M T X recovered from aspirated fluid from rabbit joints immediately following intra-articular injection of 56 mg of 18% M T X loaded P L L A 2 k microspheres  Table 3.6  156  Pharmacokinetic parameters for M T X plasma data in rabbits following an intra-articular injection of either M T X solution or M T X loaded PLLA2k microspheres (33-110 pm). The dose injected was either 1.5mg or lOmg.  The parameters are determined using WinNonlin computer  program Table 4.1  171  Intra-articular treatments to individual rabbits one day following the induction of arthritis by ovalbumin in both knee joints. PBS: 400 pL sterile PBS, M T X microspheres: 56 mg of 18% M T X loaded PLLA2k microspheres in 400 pL sterile PBS, M T X solution: 10 mg M T X solution in 400 pL sterile PBS. The dose of M T X injected was 10 mg.!  Table 4.2  196  Scoring criteria for histological analysis of rabbit joints induced with arthritis by ovalbumin and intra-articularly treated with M T X . The list of criteria was compiled and modified based on those previous reported (Kapila et al, 1995; Mould et al, 2003)  Table 4.3  203  Pharmacokinetic parameters for M T X plasma profiles of antigen (ovalbumin) induced arthritis rabbits following an intra-articular injection of either M T X solution or M T X loaded microspheres. The dose of M T X injected was lOmg. Data from normal rabbits injected with the same amount of M T X solution and M T X microspheres are shown here for ease of comparison (from Table 3.6). The parameters were determined using WinNonlin computer program  Table 4.4  212  Quantitative assessment of histopathologcial features of arthritis induced by ovalbumin in rabbit knees following M T X microspheres treatment. One day following the induction of arthritis, the right hind knees of the rabbits were injected with 56 mg of 18%) M T X loaded P L L A 2 k microspheres (33-110 pm) in 400 uL PBS. were injected with 400 uL PBS.  The left hind knee joints  Control rabbits were injected with  400pL PBS in both knee joints. The rabbits were sacrificed 14 days  following M T X treatment and knee joints were processed for histological analysis Table 4.5  229  Quantitative assessment of histopathologcial features of arthritis induced by ovalbumin in rabbit knees following M T X solution treatment. One day following the induction of arthritis, the right hind knees of the rabbits were injected with 10 mg M T X solution in 400 pL PBS. The left hind knee joints were injected with 400 pL P B S . Control rabbits were injected with 400 u,L PBS in both knee joints. The rabbits were sacrificed 14 days following M T X treatment and knee joints were processed for histological analysis  xi  230  LIST OF FIGURES Figure 1.1  A synovial joint is represented by a lateral view of the human knee  Figure 1.2  Schematic cross section of a synovial joint (top). Inset: enlargement of  5  synovial lining. Arrows indicate ultrafiltration of fluid from fenestrated capillaries into joint cavity and drainage of fluid from cavity through synovial interstitium into subsynovial space and lymphatics. (Adapted from Levick, 1995) Figure 1.3  10  Schematic representation of pathogenesis of rheumatoid arthritis. Phase 1: Antigen-presenting cell phagocytoses antigen. Phase 2: Antigen is presented to T lymphocyte. Phase 3: Activated T cell stimulates T h l and B lymphocyte production, promoting inflammation. Phase 4: Activated T cells and macrophages release factors that promote tissue destruction, increase blood flow, and result in cellular invasion of synovial tissue. A g , antigen; P M N , polymorphonuclear leukocyte; T N F - a , tumor necrosis factor-a; IFN-y, interferony; IL, interleukin, M M P , matrix metalloproteinases. (Adapted from Schuna et al., 1996)  Figure 1.4  14  Outline of the management of rheumatoid arthritis. disease-modifying antirheumatic drug; N S A I D ,  DMARD,  nonsteroidal  antiinflammatory drug; mono R x , monotherapy; combination Rx, c o m b i n a t i o n therapy. (Adapted from A m e r i c a n C o l l e g e o f Rheumatology guidelines, 2002) Figure 1.5  20  Chemical structures of A) methotrexate, B) folic acid, C) folic acid (reduced form), D) 7-hydroxy methotrexate (7-OH-MTX)  Figure 1.6  23  The chemical structures of A) poly(lactic acid) (PLA), B) poly( glycolic acid) (PGA), and C) poly(lactic-co-glycolic acid) (PLGA).  * indicates  the chiral center Figure 1.7  42  Schematic representation of A) fringed micelle model B) the chain folded model of polymer crystallinity  Figure 2.1  49  Schematic diagram of microspheres formation by the solvent evaporation method. Adapted from Watts et al (1990)  Xll  75  Figure 2.2  Particle size distributions of 10% M T X loaded P L L A microspheres manufactured from 2k, 50k, and 100k g/mole P L L A polymer using the solvent evaporation method  Figure 2.3  95  Cumulative in vitro release profiles of M T X in 0.1 M PBS at 37°C from microspheres loaded with 10% (w/w) M T X manufactured from P L L A polymers with molecular weights of 2k, 50k and 100k g/mole. Microspheres were in the size range of 33- 110 pm. Values are mean ± one standard deviation, (n=4).  (x) P L L A 2 k , (A) P L L A 5 0 k ,  (•)  PLLAlOOk Figure 2.4  98  Cumulative in vitro release profiles of M T X in 0.1M PBS at 37°C from A) microspheres loaded with various amounts of M T X manufactured from P L L A 2k g/mole. Microspheres were in the size range of 33- 110 pm.  The detailed release profile within 24 his shown in B). Values are  mean ± one standard deviation, (n=4). M T X loadings were (x) 20%(w/w), (A) 10% (w/w), (•) 5% (w/w), (4)2.5% (w/w) Figure 2.5  99  Cumulative in vitro release profiles of M T X in 0.1 M PBS at 37°C from y-irradiated (25 kGy from Co-60 source) and non-irradiated PLLA2k microspheres loaded with 10% M T X (w/w). Microspheres were in the size range of 33- 110 pm. Values are mean ± one standard deviation, (n=4). (•) non-irradiated, (•) irradiated microspheres samples  Figure 2.6  The surface morphology of A) control and B) 10% (w/w) M T X loaded P L L A 2 k microspheres  Figure 2.7  100  103  X-ray powder diffraction patterns of A ) crystalline M T X (from Handetec), B) control PLLA2k microspheres in the size range of 33-110 pm, C) 10% M T X loaded PLLA2k microspheres in the size range of 33-110 pm  Figure 2.8  104  D S C thermograms of control andl0% M T X loaded P L L A (2 kg/mole) microspheres following degradation in 0.1M PBS at 37°C. A) Control microspheres, B) 10% (w/w) M T X loaded microspheres before degradation, C) 10% M T X loaded microspheres 1 day, D) 10 % M T X loaded microspheres 28 days following degradation. Samples (3-5mg) were heated from-20°C to 200°C at 10°C /min. (Endotherms down).... 105  Figure 2.9  Molecular weight calibration curve using poly(ethylene glycol) standards.  Chromatographic conditions: Styrogel® HR3 and HR0.5 xiii  columns in series, mobile phase was THF at 1 mL/min, and the detector was a differential refractive index detector Figure 2.10  108  A) Number average molecular weight and B) weight average molecular weight profiles of control and 10% (w/w) M T X loaded P L L A 2 k microspheres in 0.1M PBS at 37°C. Microspheres were in the size range of 33- 110 pm. Values are mean ± one standard deviation, (n=3). 10% M T X loaded, (•) control microspheres  Figure 2.11  (•) 109  Cumulative weight loss profile of control and 10% M T X loaded P L L A 2 k microspheres incubated in 0.1M PBS at 37°C. Microspheres were in the size range of 33- 110 um. Values are mean ± one standard deviation, (n=3). (•) 10% M T X loaded, (•) control microspheres  Figure 2.12  110  Scanning electron micrographs of control and 10% (w/w) M T X loaded P L L A 2 k microspheres following degradation in 0.1 M PBS at 37°C. Microspheres were in the size range of 33- 110 urn. A ) Day 3, control, B) day 3, 10% M T X , C) day 7, control, D) day 7, 10% M T X  Ill  Figure 3.1  Chemical structure of aminopterin  145  Figure 3.2  Representative standard curves of A) M T X and B) 7-OH-MTX extracted from rabbit plasma and analysed by H P L C . Peak area ratio is the ratio of peak area of M T X or 7 - O H - M T X to the peak area of aminopterin measured from H P L C chromatographs.  R is the coefficient of 2  determination Figure 3.3  150  Representative standard curves of A) M T X and B) 7-OH-MTX extracted from rabbit urine and analysed by H P L C . Peak area is the area under the peak o f either M T X or 7 - O H - M T X chromatographs.  Figure 3.4  measured from  HPLC  R is coefficient of the determination 2  151  M T X concentrations in rabbit plasma following a single intra-articular injection of either M T X solution or 25mg of M T X loaded P L L A 2 k microspheres (33-110 um) in 200 uL P B S . Values are mean ± one standard deviation, (n=7). The dose of M T X injected was 1.5mg. ""Indicates statistical difference between M T X plasma concentrations of rabbits injected with M T X solution and M T X loaded microspheres by paired t-test (p<0.05).  Figure 3.5  ( • ) M T X solution, (•) M T X microspheres  159  The amount of M T X (pg) excreted at different time periods in the urine of rabbits following a single intra-articular injection of either M T X solution xiv  or 25mg of M T X loaded PLLA2k microspheres (33-110 pm) in 200uX PBS. * Indicates statistical difference between the amounts of M T X excreted from rabbits injected with M T X solution and M T X loaded microspheres by paired t-test (p<0.05). "Part of the urine sample was missed from both groups during this time period. Values are mean ± one standard deviation, (n=3). The dose of M T X injected was 1.5mg. ..160 Figure 3.6  M T X concentrations in rabbit plasma after a single intra-articular injection of either M T X solution or 56mg of 18% M T X loaded PLLA2k microspheres (33-110 pm) in 400 pX PBS. Values are mean ± one standard deviation, (n = 8 for 0-6 h period, n = 4 for 6-24 h period). The dose of M T X injected was lOmg.  * Indicates statistical difference  between M T X plasma concentrations of rabbits injected with M T X solution and M T X loaded microspheres by paired t-test (p<0.05). M T X solution, (•) M T X microspheres Figure 3.7  (•) 165  7 - O H - M T X concentrations in rabbit plasma following a single intra-articular injection of lOmg M T X solution in 400 uX PBS. (•) rabbit #15, (•) rabbit #5, (A) the remaining animals in the group, values are mean ± one standard deviation (n = 6)  Figure 3.8  166  7 - O H - M T X concentrations in rabbit plasma following a single intra-articular injection of 56 mg of 18% M T X loaded P L L A 2 k microspheres (33-110 pm) in 400 uX PBS. The injections provided a 10 mg M T X dose.  (A) rabbit # 14,(H) rabbit # 4, (•) the remaining animals  in the group, values are mean ± one standard deviation (n = 6) Figure 3.9  167  The amount of A ) M T X (pg) and B) 7 - O H - M T X (pg) excreted at different time periods in the urine of rabbits following a single intra-articular injection of either 10 mg M T X solution in 400 uX PBS or 56mg of 18% M T X loaded PLLA2k microspheres (33-110 pm) in 400 uX PBS. Values are mean ± one standard deviation, (n = 8 for the period of 0-6 h, n= 4 for the period of 6-24 h). The dose of M T X injected was lOmg.  * Indicates statistically different between two treatments.by  paired r-test (p<0.05) Figure 3.10  168  The amount of M T X in A) pool fluid aspirated from synovial joints and B) the synovial tissues, 6 h and 24 h following a single intra-articular injection of either 10 mg M T X solution in 400 uX PBS or 56mg of 18% M T X loaded microspheres in 400 uX PBS. The values reported in B) are XV  corrected to account 100% M T X recovery from synovial tissues. Values are mean + one standard deviation, (n = 4). The dose of M T X injected was lOmg. * Indicates statistically different between two treatments by paired t-test (p<0.05) Figure 3.11  169  Histological analysis of synovial tissues following intra-articular injection of 56 mg of 18%> M T X loaded P L L A 2 k microspheres (33-110 urn) and sacrificed at 6 h following injection. (A) synovial tissue with dark color granules visible in the rectangular area (40x magnification), (B) higher magnification of the rectangular area (400x magnification).. 173  Figure 3.12  Histological analysis of synovial tissues following intra-articular injection of A) 10 mg M T X solution in 400 uL PBS and B) 56 mg of 18% M T X loaded P L L A 2 k microspheres ( 33-110 urn) and sacrificed at 24 h following injection. (Magnification 100 x). Both graphs showed no evidence of synovial proliferation as indicated by the arrows  Figure 3.13  174  The A S T activities in rabbit plasma after a single intra-articular injection of either 10 mg M T X solution in 400 uL PBS or 56mg of 18% M T X loaded P L L A 2 k microspheres (33-110 urn) in 400 uL PBS. Values are mean ± one standard deviation, (n = 8 for 0-6 h period, n = 4 for 6-24 h period). The dose of M T X injected was 10 mg  Figure 4.1  176  The gross appearance of a decalcified rabbit knee joint. A) Rabbit knee joint transected 3-4 cm above and below the knee joint. B) The joint was bisected in the sagittal plane. C) The gross appearance of the knee joint following bisection  Figure 4.2  201  M T X concentrations in plasma of antigen (ovalbumin) induced arthritis rabbits following a single intra-articular injection of either 10 mg M T X solution in 400 p L PBS or 56 mg of 18% P L L A 2 k M T X loaded microspheres (33-110 pm) in 400 uL PBS into the right hind knee joint. Left hind knee joints were injected with 400 pL PBS.  Values are mean  ± one standard deviation, (n = 5). The dose of M T X injected was lOmg. (•) M T X solution, (•) M T X microspheres Figure 4.3  207  M T X plasma concentrations of individual antigen (ovalbumin) induced arthritis rabbits following a single intra-articular injection of 10 mg M T X solution in 400 pL PBS into the right hind knee joint. Left hind knee  xvi  joints were injected with 400 uX PBS.  (•) rabbit #3, (•) rabbit #5, (A)  rabbit #7, (x) rabbit #9, (•) rabbit #11 Figure 4.4  208  M T X plasma concentrations of individual antigen (ovalbumin) induced arthritis rabbits following a single intra-articular injection of 56 mg of 18% M T X loaded P L L A 2 k microspheres (33-110 pm) in 400 pX PBS into the right hind knee joint. Left hind knee joints were injected with 400 pX PBS.  The dose of M T X injected was 10 mg M T X .  #2, (•) rabbit #4, (A) rabbit #6, (x) rabbit #10, (•) rabbit #12 Figure 4.5  (•) rabbit 209  7-OH-MTX plasma concentrations of individual antigen (ovalbumin) induced arthritis rabbits following a single intra-articular injection of 10 mg M T X solution in 400 uX PBS into the right hind knee joint. Left hind knee joints were injected with 400 pX PBS.  (•) rabbit #3, (•)  rabbit #5, (A) rabbit #7, (x) rabbit #9, (•) rabbit #11 Figure 4.6  210  7-OH-MTX plasma concentrations of individual antigen (ovalbumin) induced arthritis rabbits following a single intra-articular injection of 56 mg of 18% M T X loaded PLLA2k microspheres in 400 pX PBS to the right hind knee joint. Left hind knee joints were injected with 400 uX PBS.  The dose of M T X injected was 10 mg M T X .  (•) rabbit #2, (•)  rabbit #4, (A) rabbit #6, (x) rabbit #10, (•) rabbit #12 Figure 4.7  211  Knee joint swelling following antigen (ovalbumin) induction of arthritis in rabbits. One day following the induction of arthritis, the right hind knees of the rabbits were injected with either 10 mg M T X solution in 400 pX PBS or 56 mg of 18% M T X loaded microspheres (33-110 pm) in 400 uX PBS.  The left hind knee joints were injected with 400pX PBS. A)  Comparison of knee joint swelling between M T X solution treated knees and 400pL PBS treated contralateral knees, B) comparison of knee joint swelling between M T X loaded microspheres treated and PBS treated contralateral knees, C) comparison of knee joint swelling between M T X solution treated knees and M T X microspheres treated knees. Values are mean + one standard deviation, (n = 5). The dose of M T X injected was lOmg. (•) M T X solution, (•) M T X microspheres, (x) PBS Figure 4.8  215  Optical micrographs of the cartilage of a normal rabbit knee joint obtained from another independent study, provided by the Faculty of Medicine, University, of Calgary. A) Cartilage of femur, tibia and part of the meniscus of the knee joint. (25x), B).Cartilage of tibia at a higher xvii  magnification (lOOx). Zones of cartilage are marked. Round circular cells in dark purple color are the chondrocytes. Figure 4.9  220  Optical micrographs of the synovial membrane of a normal rabbit knee joint obtained from another independent study, provided by the Faculty of Medicine, University of Calgary.  A ) Synovial membrane (25x)  (indicated by the arrows), B) Synovial membrane at a high magnification (lOOx). car: cartilage, adp: adipose cells, j cav: joint cavity Figure 4.10  221  Representative micrographs of cartilage destruction due to arthritis induced by ovalbumin in rabbit knee joints. One day following the induction of arthritis, the right hind knees of the rabbits were injected with either lOmg M T X solution in 400 pL PBS or 56 mg of 18% M T X loaded P L L A 2 k microspheres (33-110 um) in 400 uL PBS. The left hind knee joints were injected with 400 pL PBS. The rabbits were sacrificed 14 days following M T X treatment and knee joints were processed for histological analysis. A) Minimal abrasion as indicated by the arrow on the surface of the cartilage represents score 1. The joint was injected with 400 uL PBS.  B) Destruction with superficial loss of  chondrocytes and cartilage disruption indicated by the arrow (score 2). The joint was injected with M T X microspheres.  C) Moderate loss of  chondrocytes and cartilage disruption to mid zone (score 3). The joint was injected with PBS.  D) Destruction with severe loss of chondrocyte  to tide mark as indicated by the arrow (score 4). The joint was injected with PBS.  A l l micrographs shown are at a magnification of 25x.  meniscus, car: cartilage, j cav: joint cavity Figure 4.11  men: 222  Representative micrographs of cartilage interstitial loss of matrix due to arthritis induced by ovalbumin in rabbit knee joints. One day following the induction of arthritis, the right hind knees of the rabbits were injected with either 10 mg M T X solution in 400pL PBS or 56 mg of 18% M T X loaded P L L A 2 k microspheres (33-110 um) in 400 pL PBS. The left hind knee joints were injected with 400 pL PBS. The rabbits were sacrificed 14 days following M T X treatment and knee joints were processed for histological analysis. No scores were given for this histopathological feature.  A) interstitial matrix loss (pointed by arrows)  but surface remained intact. The joint was injected with M T X solution. B) Interstitial matrix loss accompanied by abnormal cellularity indicated  xviii  by the rectangle. The joint was injected with M T X microspheres. C) The rectangular section of B with a higher magnification (100 x). The interstitial matrix loss was evidenced by the decrease of intensity of the color stain compared to intensity of stained cartilage in Figure 4.8. men: meniscus, car: cartilage, j cav: joint cavity Figure 4.12  223  Representative micrographs of pannus formation due to arthritis induced by ovalbumin in rabbit knee joints. One day following the induction of arthritis, the right hind knees of the rabbits were injected with either 10 mg M T X solution in 400 uL PBS or 56 mg of 18% M T X loaded PLLA2k microspheres (33-110 pm) in 400 uX PBS. were injected with 400 pX PBS.  The left hind knee joints  The rabbits were sacrificed 14 days  following M T X treatment and knee joints were processed for histological analysis. A ) Pannus with superficial cartilage destruction as indicated by the rectangle represents score 1. The joint was injected with PBS. A higher magnification (lOOx) of the rectangular section is shown in B). C) Pannus with destruction to the depth of mid zone as indicated by the rectangle (score 2). The joint was treated with M T X microspheres. The cut parallel to the surface of the cartilage as shown by the arrow is an artifact due to histological processing. A higher magnification (lOOx) of the rectangular section is shown in D). men: meniscus, j cav: joint cavity Figure 4.13  224  Representative micrographs of synovial tissue hyperplasia due to arthritis induced by ovalbumin in rabbit knee joints. One day following the induction of arthritis, the right hind knees of the rabbits were injected with either 10 mg M T X solution in 400 uX PBS or 56 mg of 18% M T X loaded P L L A 2 k microspheres (33-110 pm) in 400 pX PBS. The left hind knee joints were injected with 400 pX PBS. The rabbits were sacrificed 14 days following M T X treatment and knee joints were processed for histological analysis. A) 4-6 layer increase in synovial thickness as indicated by the arrow represents score 1. The joint was injected with P B S . B) >7 layer increase in synovial thickness as indicated by the double arrow (score 2). The joint was injected with PBS. Both A ) and B) are at a magnification of 25x.  Figure 4.14  car: cartilage  225  Representative micrographs of villous hyperplasia in synovial tissues due to arthritis induced by ovalbumin in rabbit knee joints. One day  xix  following the induction of arthritis, the right hind knees of the rabbits were injected with either 10 mg M T X solution in 400 uX PBS or 56 mg of 18% M T X loaded P L L A 2 k microspheres (33-110 pm) in 400 uX PBS. The left hind knee joints were injected with 400 uX PBS. The rabbits were sacrificed 14 days following M T X treatment and knee joints were processed for histological analysis. A) A few scattered and short villous hyperplasia as indicated by the arrow represents score 1. The joint was injected with PBS. B) Marked finger like villi indicated by the arrow (score 2).  The joint was injected with M T X microspheres.  micrographs are at a magnification of 25 x. j cav: joint cavity  Both 226  Figure 4.15 Representative micrographs of inflammatory cellular infiltration in synovial tissues due to arthritis induced by ovalbumin in rabbit knee joints. One day following the induction of arthritis, the right hind knees of the rabbits were injected with either 10 mg M T X solution in 400 pL PBS or 56 mg of 18% M T X loaded microspheres in 400 pX PBS. The left hind knee joints were injected with 400 uX PBS. The rabbits were sacrificed 14 days following M T X treatment and knee joints were processed for histological analysis. A ) Increased cellularity and fibrin (pink stain) in the subsynovial layer of the synovial tissue represents score 3. The joint was injected with PBS. Magnification 25x. A higher magnification (lOOx) of the region of increased cellularity is shown in B)  227  Figure 4.16 Representative micrographs of inflammatory exudates in the joint cavity due to arthritis induced by ovalbumin in rabbit knee joints. One day following the induction of arthritis, the right hind knees of the rabbits were injected with either 10 mg M T X solution in 400 pX PBS or 56 mg of 18% M T X loaded microspheres in 400 uX PBS. The left hind knee joints were injected with 400 uX PBS. The rabbits were sacrificed 14 days following M T X treatment and knee joints were processed for histological analysis. A ) Exudates in the joint cavity represent score 1. The joint was treated with M T X solution. (Magnification 25 x). A higher magnification (100 x) of the exudates is shown in B ) . inflammatory exudates, j cav: joint cavity  XX  exu: 228  LIST OF ABBREVIATIONS r)o  Viscosity of the solvent  •n i  Relative viscosity  rired  Reduced viscosity  r| p S  Specific viscosity  [r|]  Intrinsic viscosity  AHf  Heat of fusion  AHc  Hear of recrystallization  ANOVA  Analysis of variance  AST  Aspartamine transferase  AUC  Area under the curve  AUMC  Area under the moment curve  C  Concentration  D  Diffusion coefficient  Do  Dose administered  DMARD  Disease-modifying anti-rheumatic drug  DSC  Differential scanning calorimetry  GPC  Gel permeation chromatography  HPLC  High performance liquid chromatography  IFN  Interferon  IL  interleukin  J  Flux  Mn  Number-average molecular weight  Mv  Viscosity-average molecular weight  Mw  Weight-average molecular weight  MMP  Matrix metalloproteinases  MRT  Mean residence time  MTX  Methotrexate  NSAID  Non-steroidal anti-inflammatory drugs  PBS  Phosphate buffered saline  PDLLA  Poly(D,L-lactic acid)  PGA  Poly(glycolic acid)  PLA  Poly(lactic acid)  PLLA  Poly(L-lactic acid)  PLGA  Poly(lactic-co-glycolic acid)  re  xxi  PMN  Polymorphonuclear leukocytes  rpm  Revolutions per minute  SD  Standard deviation  k  1 order rate constant for non-compartmental analysis  k  st  a  Absorption constant for one-compartmental model  ke  Elimination rate constant for one-compartmental model  t  Time  ti/2  elimination half-life  ti/2(abs)  Absorption half-life  Tc  Recrystallization temperature  Tg  Glass transition temperature  Tm  Melting point  TNF  Tumor-necrosis factor  UV  Ultraviolet  V  Volume of distribution  Xc  Degree of crystallinity  XRD  X-ray diffraction  xxii  ACKNOWLEDGEMENTS This thesis could not be complete without the help of many people.  First, I would  like to thank my supervisor, Dr. Helen Burt, for her high academic standards, guidance and support over the course of my Ph.D. years.  I would also like to thank my committee  members for their advice and direction on my research project:  Drs. John McNeill, Don  Brooks, Kishor Wasan, and Don Lyster. Thank you to Mr. John Jackson for always being there to solve any problems in the lab and to provide support when I was frustrated with my research.  Thank you to all my  colleagues for their friendship and help: Karen, Kevin, Jason, Ruiwen, Chris, Tobi, Wes, and Chiming. I would like to thank Mr. Michael Boyd and Ms. Vera Risovic for helping me with the pharmacokinetic study.  I would also like to thank Drs. David Hart and Paul Salo and  Ms. Carol Reno and Ms. Ruth Seerattan from the University of Calgary for their great help and valuable discussions in the in vivo efficacy study. Financially support from the Canadian Arthritis Network, the Canadian Institutes of Health Research, and University Graduate Fellowship is gratefully acknowledged. I would like to dedicate this work to my parents. Thank you for your unconditional love.  xxiii  Chapter 1  PROJECT OVERVIEW AND BACKGROUND 1.1  PROJECT OVERVIEW  Rheumatoid arthritis is an autoimmune disease characterized by progressive and irreversible damage of the syno vial-lined joints, resulting in the loss of joint space, a decrease in joint function and ultimately, deformity. Despite an increased understanding of the pathophysiology of rheumatoid arthritis, there are no treatments that cure rheumatoid arthritis. Therefore, the therapeutic goals are a remission of symptoms involving the joints, a return of full function, and the maintenance of remission with disease-modifying anti-rheumatic drug (DMARD) therapy (O'Dell et ah, 2004). Methotrexate (MTX) is one of the most frequently used D M A R D s .  Although the  exact mechanism of action is still not fully understood, the efficacy of M T X is related to its cytotoxic and anti-inflammatory effects (Furst et al., 1990). When administered in low weekly oral doses, M T X effectively suppresses inflammation in rheumatoid arthritis. However, oral use of the drug is limited by its systemic toxic effects.  Therefore,  attempts have been made to improve the efficacy and reduce toxicity by administering the drug intra-articularly. These outcomes were not achieved due to rapid removal of the drug from the joint cavity (Wigginton et al., 1980).  1  Intra-articular injection of a controlled release drug delivery system has been proposed in order to maintain an effective concentration of the drug in the joint cavity. Several studies (Nishide et al., 1999; Ramesh et ah, 1999; Horisawa et al, 2002a; Liggins et al., 2004) have shown that intra-articular microsphere formulations composed of biodegradable and biocompatible polymers are well tolerated in rabbit joints. Inflammatory responses were not significant, although a temporary increase in cell infiltration into the synovial fluid has been reported.  However, very few studies have  been carried out to study the pharmacokinetics and efficacy of a D M A R D encapsulated controlled release system for intra-articular treatment of rheumatoid arthritis. The aim of this work was to develop a M T X loaded polymeric microspheres formulation suitable for the potential intra-articular treatment of rheumatoid arthritis. The first part of the thesis focuses on the formulation and characterization of M T X encapsulated poly(L-lactic acid) microspheres.  The second and major part of the thesis  evaluates the pharmacokinetics of M T X in healthy rabbits and the last part describes an in vivo efficacy study of M T X encapsulated microspheres following intra-articular delivery in a rabbit model of antigen induced arthritis.  1.2 1.2.1  SYNOVIAL JOINT STRUCTURE Synovial joint anatomy  Synovial joints are freely movable joints that have a joint cavity, and ligaments help to support the articulating bones.  The functions of synovial joints are to provide a wide  range of precise, smooth movements, at the same time maintaining stability, strength and, in certain aspects, rigidity in the body (van den Graaff, 1995). The structure of a synovial joint is shown in Figure 1.1.  Synovial joints are  enclosed by a fibroelastic joint capsule that is filled with lubricating synovial fluid. Synovial fluid is confined within the cavity by a thin sheet of tissue called synovium (also known as synovial membrane or synovial lining).  Synovium generates the synovial  fluid and supplies it with oxygen and nutrients for transmission to the cartilage. It achieves this by virtue of its rich microcirculation and high blood flow.  Around the  margins of the joint cavity, small folds of synovium with or without blood vessels sometimes project into the cavity and these folds are called " v i l l i " (Levick, 1995). The bones that articulate in a synovial joint are capped with a smooth articular cartilage. The avascular articular cartilage depends on the alternating compression and decompression during joint activity for the exchange of nutrients and waste products with the synovial  3  fluid.  Tough, fibrous cartilaginous pads called menisci are located within the capsule of  certain synovial joints and serve to cushion, as well as to guide, the articulating bones.  4  Suprapatellar bursa  Synovial membrane Femur Tendon of quadriceps femoris muscle Patella Synovial membrane fi  jjjli— Subcutaneous prepartellar bursa  Articular cartilage Infrapatellar fat pad Meniscus Subcutaneous infrapatellar bursa Joint cavity filled with synovial fluid  Infrapatellar bursa  Patellar ligament  Tibia  Figure 1.1  A synovial joint is represented by a lateral view of the human knee  joint. Adapted from Human Anatomy (Adapted from van den Graaff, 1995)  5  1.2.2  Synovium  ,  The synovium itself is very thin, typically around 15-20 urn thick in the rabbit knee (Knight and Levick, 1983; Levick and McDonald, 1989) and 60 pm thick in the human knee (Stevens et al., 1991).  Synovium backs onto a broader zone of loose connective  tissue and fat cells called subsynovium, where terminal arteries, veins, and a network of terminal lymphatic plexus drains away fluid and macromolecules that have seeped out through the synovial lining (Figure 1.2) (Simkin and Benedict, 1990; Jensen et al., 1993; Levick, 1995) Two main types of cells are found in the synovial lining.  Type A cells contain a  prominent Golgi complex and many vesicles, but little rough endoplasmic reticulum. They are also more endowed with cell processes, mitochondria, intracytoplasmic filament, and lysosomes (Ghadially, 1983).  Type B cells are well endowed with rough  endoplasmic reticulum but with little Golgi complex, vesicles and vacuoles (Ghadially, 1983).  Type A cells are often called macrophage-like cells because of the paucity of  rough endoplasmic reticulum.  The main functions of type A cells are to phagocytose  and to secrete hyaluronic acid from Golgi complexes (Ghadially, 1983; Brown et al., 1991).  Type B cells are often called fibroblast-like cells. The main function of type B  cells is to produce a protein-rich secretion.  The secretion may include procollagen,  6  collagenase, and lubricin, a lubricating glycoprotein found in the synovial fluid (Fraser et al., 1977). Type B cells also have phagocytotic ability when stimulated (Ghadially et al., 1982). Senda and coworkers (1999) showed that both type A and B synovial cells possessed the ability to phagocytose latex particles with a diameter of 240 nm, and that such activity was more intense in type A cells. Normal synovium possesses a row of capillaries at about 5 pm (rabbit knee) or about 30 urn (human knee) beneath the surface.  Many of the capillaries bear fenestrations  which are often on the side facing the joint cavity (Levick and McDonald, 1995). The high capillary density, superficial location, and orientation of the fenestrations are well adapted for synovial fluid formation and nutrient supply.  1.2.3  A r t i c u l a r cartilage  Articular cartilage is an avascular, alymphatic, aneural tissue which covers the articular ends of bone.  It is composed of cells (chondrocytes) set in an abundant matrix.  The chondrocytes occupy only about 0.01 to 0.1% of the volume of the tissue (Ghadially et al., 1982). The matrix contains collagen fibrils, proteoglycans and a number of organic and inorganic solutes.  The functions of articular cartilage include transmission  and distribution of high loads to underlying bone, maintenance of contact stresses at  7  acceptably low levels, movement with little friction, and shock absorption (Ghadially et al, 1982). The margin of articular cartilage blends gradually with the synovial membrane and periosteum.  The thickness of articular cartilage varies from joint to joint, from one area  of the joint to another in the same joint, and from species to species.  The thicknesses of  articular cartilage in the human hip and knee range from about 2 to 4 mm (Meachim and Stockwell, 1979) and the cartilage covering the femoral condyle of the rabbit is a little under 0.5 mm thick (Davies et al., 1962; Ghadially et al., 1982).  1.2.4  Synovial fluid  Synovial fluid is a slippery, viscoelastic liquid that fills joint cavity.  It provides  freedom of movement, lubricates the joint, and acts as a diffusional bridge and convective transport medium for nutrients and waste products (Levick, 1995). Synovial fluid formation begins with the passive ultrafiltration of plasma across the superficial, surface oriented fenestrations of the synovial capillaries. The ultrafiltrate flows through the interstitium, between the synovial lining cells to enter the joint cavity and becomes synovial fluid as lubricating macromolecules, hyaluronic acid and lubricin, are actively secreted into it by the synovial cells (Knox et al., 1988).  Synovial fluid is not a static  pool, but is continually being absorbed and replenished by the synovial lining of the joint  8  cavity. As shown in Figure 1.2, synovial fluid and macromolecules drain from the joint cavity via the interstitial path between the lining cells to reach the lymphatic capillary network at the border between synovium and subsynovium (Levick, 1995; Levick and McDonald, 1995). The macromolecules are returned to the circulation by the lymphatic system (Simkin and Benedict, 1990; Jensen, et al., 1993). The synovial fluid volume in normal human knees is less than 2 mL (Netter et al., 1989) while the synovial fluid volume aspirated in rheumatoid arthritis and osteoarthritis patients is between 20-40 mL (Hunter and Blyth, 1999). The rate of volume turnover of synovial fluid is estimated to be of the order of 3-4 pL/h/cm synovium in rabbit and human knees and this corresponds 2  to a turnover time of roughly 1 h (Levick, 1987). In rheumatoid arthritis, the rate of turnover of synovial fluid increases 4 fold, while the fluid volume increases more than 8 fold, so the turnover time (i.e., volume/turnover rate) actually lengthens (Simkin, 1979; Levick, 1995)  9  Collagenous capsule  Articular cartilage  Joint cavity  Synovium Fenestrated capillary Loose areolar connective tissue  Figure 1.2 Schematic cross section of a synovial joint (top). Inset: enlargement of synovial lining.  Arrows indicate ultrafiltration of fluid from fenestrated capillaries into  joint cavity and drainage of fluid from cavity through synovial interstitium into subsynovial space and lymphatics. (Adapted from Levick, 1995)  10  1.3  RHEUMATOID ARTHRITIS  Rheumatoid arthritis is a chronic inflammatory disease of unknown etiology and complex multifactorial pathogenesis, affecting joints and other tissues. - It affects up to 1-3% of the population and is two to three times more frequent in women than in men (Grassi et al., 1998; Howe, 1998). The disease can start at any age, with a peak incidence between the fourth and the sixth decade of life (Smolen and Steiner, 2003).  1.3.1  Pathogenesis  Although the factors that initiate the inflammatory process of rheumatoid arthritis are unknown, many pathways involved in the generation of the disease have been recognized.  The pathogenesis of rheumatoid arthritis has been reviewed by Hasunuma  et al (1998), van den Berg and Bresnihan (1999), Bingham (2002), and Smolen and Steiner (2003).  Rheumatoid arthritis is generally considered to be an autoimmune  disease, with primary or secondary involvement of T lymphocytes (van den Berg and Bresnihan, 1999). It has been speculated that rheumatoid arthritis could be triggered by infectious agents, but proof of this is still lacking (Smolen and Steiner, 2003). Inflammation and tissue destruction of the joint in rheumatoid arthritis results from a T lymphocyte response, triggered by the binding of foreign antigens to receptors on T lymphocytes (Schuna et al., 1996). T lymphocytes can undergo polarization into either  11  Thl cells or Th2 cells which can be mutually inhibitory (Smolen and Steiner, 2003). Thl cells mainly secrete pro-inflammatory cytokines such as interferon-y (IFN-y) and tumor-necrosis factor-(3 (TNF-P); whereas Th2 cells mainly secrete anti-inflammatory cytokines such as interleukin-4 (IL-4), IL-5, IL-13, and IL-10 (Romagnani, et ah, 2003). The T lymphocytes infiltrating the synovial membrane in rheumatoid arthritis are primarily CD4+ memory cells that clearly have a Thl bias (Smolen and Steiner, 2003). These T cells, by cell-cell contacts, or by different cytokines, such as IFN-y, TNF-cc and IL-17, activate monocytes, macrophages and synovial fibroblasts (Stout, 1993; Smolen and Steiner, 2003).  These cells then overproduce pro-inflammatory cytokines, mainly  TNF-a, IL-1, and IL-6 which seem to be the pivotal cytokines that lead to chronic inflammation (Firestein et al, 1990; Smolen and Steiner, 2003).  These  pro-inflammatory cytokines, once engaged to their receptors on specific cell surfaces, trigger various signal transduction cascades which lead to the induction of genes whose products mediate inflammation and tissue degradation (Bingham, 2002).  These  products include various cytokines, chemokines, and tissue-degrading enzymes, such as the matrix metalloproteinases (Bingham, 2002; Smolen and Steiner, 2003).  The  endogenous anti-inflammatory agents, such as soluble cytokine receptors and receptor antagonists, anti-inflammatory cytokines, or regulatory T lymphocytes, are insufficient to  12  counterbalance the cascade of pro-inflammatory events in chronic inflammation (Bingham, 2002; Smolen and Steiner, 2003). B lymphocytes are also activated by T lymphocytes. Activated B lymphocytes produce plasma cells, which form antibodies.  These antibodies, in combination with  complement, result in the accumulation of polymorphonuclear leukocytes (PMNs). These PMNs release cytotoxins, free oxygen radicals, and hydroxyl radicals, which promote cellular damage to synovium and bone (Schuna et al., 1996; Howe, 1998; van den Berg and Bresnihan, 1999). Overall, many cell populations are involved in rheumatoid arthritis pathogenesis. As a result of inflammatory infiltration, T lymphocytes, B lymphocytes, and plasma cells are found in greatly increased numbers in the subsynovial layer (Bingham, 2002). The synovial membrane contains type A and type B synovial cells which proliferate resulting in an aggressive tissue called pannus at the cartilage-bone junction (Bingham, 2002; Smolen and Steiner, 2003). Pannus contains matrix metalloproteinases and osteoclasts that invade the cartilage and eventually the bone surface, producing erosions of bone and cartilage, leading to destruction of the joint (Smolen and Steiner, 2003). A simplified schematic representation of the pathogenesis of rheumatoid arthritis is shown in Figure 1.3.  13  Macrophages IFN-y, T N F - a ,  Prostaglandins,  and IL-17 Cytotoxins,  ®  TNF-a  Cytokines, Cytotoxins,  Inflammation, Synovial proliferation,  TNF-a Cartilage and bone destruction Collagenase, MMPs, Cytotoxins j Antibodies Macrophages  Lymphocyte  Complement Plasma cell PMN  Figure 1.3  Schematic representation of pathogenesis of rheumatoid arthritis. Phase 1:  Antigen-presenting cell phagocytoses antigen. Phase 2: Antigen is presented to T lymphocyte. Phase 3: Activated T cell stimulates T h l and B lymphocyte production, promoting inflammation. Phase 4: Activated T cells and macrophages release factors that promote tissue destruction, increase blood flow, and result in cellular invasion of synovial tissue.  Ag, antigen; P M N , polymorphonuclear leukocyte; TNF-a, tumor  necrosis factor-a; IFN-y, interferony; IL, interleukin, M M P , matrix metalloproteinases. (Adapted from Schuna et al., 1996)  14  1.3.2  Clinical presentation  The clinical features of rheumatoid arthritis can be classified as articular and extra-articular.  Rheumatoid arthritis is a polyarthritis. It involves many joints (six or  more), although in the early stages of the disease, only one or a few joints might be afflicted.  Virtually all peripheral joints can be affected by the disease; however, the  most commonly involved joints are those of the hands, feet and knees (Smolen et ah, 1995), and distal interphalangeal joints are usually spared. and functional limitation are typical of active disease.  Pain, early morning stiffness  The onset is usually insidious and  the joints become involved in an additive and progressive manner.  Extra-articular  symptoms include constitutional symptoms and involvement of other systems that range from rheumatoid nodules to life- threatening vasculitis (Schuna, et al., 1996).  1.3.3  Management of the disease  Medications that are used to treat rheumatoid arthritis are divided into three main classes: nonsteroidal anti-inflammatory drugs (NSAIDs), corticosteroids and disease modifying anti-rheumatic drugs (DMARDs).  According to the guidelines for the  management of rheumatoid arthritis published by the American College of Rheumatology in 2002, D M A R D therapy should be started within 3 month of the diagnosis of rheumatoid arthritis and corticosteroid and NSAIDs may be used for pain  15  relief and anti-inflammatory purposes (Figure 1.4).  There are no treatments that cure  rheumatoid arthritis; therefore, the therapeutic goals are a remission of symptoms involving the joints, a return of full function, and the maintenance of remission with D M A R D therapy (O'Dell, 2004). 1.3.3.1  Nonsteroidal anti-inflammatory drugs  NSAIDs are particularly helpful during the first few weeks in which a patient has symptoms, because the drugs provide partial relief of pain and stiffness until a definitive diagnosis of rheumatoid arthritis can be established.  NSAIDs have not been shown to  slow the progression of the disease; therefore, in long-term care, NSAIDs should be used together with D M A R D s (American College of Rheumatology, 2002; O'Dell, 2004). Long-term administration of NSAIDs may result in gastrointestinal ulcers, perforation, and hemorrhage (Howe, 1998).  Cyclooxygenase-2 (COX-2) inhibitors, which decrease  the incidence of gastric and duodenal ulcers by approximately 50 percent as compared with traditional NSAIDs, have been introduced (Bombardier et al., 2000; Silverstein, et al., 2000; Fitzgerald and Patrono, 2001).  However, the efficacy of the COX-2 inhibitors  is no better than that of the older and less expensive NSAIDs (Fitzgerald and Patrono 2001).  It has been found that when used in high doses, COX-2 inhibitors may increase  the risk of cardiovascular diseases in patients (Maxwell and Webb, 2005).  16  1.3.3.2  Corticosteroids  Corticosteroids are potent suppressors of the inflammatory response in rheumatoid arthritis and in many other diseases.  Studies have shown that corticosteroids decrease  the progression of rheumatoid arthritis, detected using radiographic techniques (Kirwan, 1995; Hickling et al., 1998; van Everdingen et al., 2002).  Oral corticosteroids can be  used in "bridging" therapy, continuous low-dose therapy, or short-term high-dose therapy to control flares of rheumatoid arthritis. In terms of bridge therapy, oral corticosteroids can be used to control pain and synovitis while D M A R D s are taking effect (O'Dell, 2004). Adverse effects are the major limitations to the use of corticosteroids. They include hypothalamic-pituitary axis suppression, Cushing's syndrome, osteoporosis, myopathies, glaucoma, cataracts, gastritis, hypertension, hirsuitism, electrolyte imbalance, glucose intolerance, skin atrophy, and increased susceptibility to infections (Schuna  a/., 1996).  1.3.3.3 Disease modifying anti-rheumatic drugs Optimal management of rheumatoid arthritis requires rapid and sustained suppression of inflammation with D M A R D s , which are defined as medications that retard or halt the progression of disease.  Disease modification is most convincingly  demonstrated by the ability of the medications to decrease radiographic progression. A meta-analysis of blinded clinical trials has suggested that the relative efficacies of 17  methotrexate (MTX), sulfasalazine, intramuscular gold, and penicillamine are similar (Felson et al., 1990; O'Dell, 2004). Antimalarial drugs (e.g., chloroquine and hydroxychloroquine) are less effective.  Penicillamine, because of concern about its  toxicity, and oral gold, because of its marginal efficacy (Felson et al., 1990), are rarely used today (Mikuls and O'Dell, 2000). However, all presently used D M A R D s have limited efficacy, toxicity problems, or both. Drug retention rates in clinical practice are relatively low.  Within 1-2 years, the majority of patients have to stop a given D M A R D  course, with the exception of M T X , which has a median retention rate of 3-4 years (Smolen and Steiner, 2003). M T X is most often selected for initial therapy (Mikuls and O'Dell, 2000). It has demonstrated efficacy and durability, a long-term track record of acceptable toxicity, and low cost (Ortendahl et al, 2002). 1.3.3.4 Biologic disease modifying anti-rheumatic drugs Cytokines play important roles in the progression of rheumatoid arthritis. Currently, there are three biologic products available to treat rheumatoid arthritis. Infliximab, etanercept, and dalimumab inhibit the actions of TNF-a, and anakinra inhibits the action of IL-1 (Olsen and Stein, 2004). These biologic D M A R D s have significantly improved the clinical response and radiographic progression of patients when used alone or in combination with M T X , compared to either placebo or M T X plus  18  placebo therapies (O'Dell, 2004).  Currently, biologic D M A R D s have been approved  for use alone, or in combination with M T X , when M T X itself provides suboptimal therapeutic effects.  The adverse effects of these agents include injection-site and  infusion reactions, autoimmune responses and increase in infections (Olsen and Stein, 2004).  19  Establish Diagnosis of Rheumatoid Arthritis Early Document Baseline Disease Activity and Damage C Estimate Prognosis i"  Initiate Therapy • Patient Education • Start DMARD(s) Within 3 Months • Consider NSAID • Consider Local or Low-Dose Systemic Steroids • Physical Therapy/Occupational Therapy  1  1  y  Periodically Assess Disease Activity  Inadequate Response (i.e., ongoing active disease after 3 months Of maximal therapy)  Adequate Response with Decreased Disease Activity  Change/Add DMARDs (  MTX Naive  MTX  MX Other Combination MonoRx Rx  Suboptimal MTX Response  ^  Combination Rx  i  X  Other MonoRx  Biologies  Mono Rx  Combination Rx  Multiple DMARD Failure  Symptomatic and/or Structural Joint Damage  Surgery  Figure 1.4  Outline of the management of rheumatoid arthritis.  DMARD,  disease-modifying antirheumatic drug; NSAID, nonsteroidal antiinflammatory drug; mono Rx, monotherapy; combination Rx, combination therapy. (Adapted from American College of Rheumatology guidelines, 2002)  20  1.4 1.4.1  METHOTREXATE Chemistry  Methotrexate (MTX) (Figure 1.5) is N - [4- {{(2,4-diamino-6-pteridinyl)-methyl} methyl-amino} benzoyl] glutamic acid. M T X is also known as 4-amino-10-methylfolic acid and amethopterin.  The chemical  formula of M T X is C20H22N8O5 and the molecular weight of anhydrous M T X is 454.46 g/mole (Budavari, 1996).  M T X is a folic acid analog.  There are three main regions in  its structure as follows (1) pteridine ring, (2) p-aminobenzoic acid (the bridge region), and (3) glutamic acid (Figure 1.5) (Rahman and Chhabra, 1988). M T X is a bright yellow-orange, odorless crystalline powder.  Melting of M T X is  accompanied by decomposition in the temperature range 185-204°C (Budavari, 1996). The molecule is negatively charged at neutral pH with pKa values of 2.15 and 3.8 and therefore its aqueous solubility is pH dependent (Hansen et al, 1983).  The solubility of  M T X in double distilled water at 20°C was reported be 50 pg/mL (Chan and Gonda, 1991}. The aqueous solubility of M T X is the lowest at pH 3 (Hansen et al, 1983). At pH below 2.5, a slight increase in solubility is observed due to the increased ionization of the M T X molecule in this region (Fort et al, 1990).  M T X is practically insoluble in  alcohol, chloroform, ether and dichloromethane (Gennaro et al, 1995).  21  At a constant pH and temperature, the degradation of M T X solutions display a first-order kinetic behavior (Chatterji, et al., 1978; Hansen et al., 1983).  The  degradation of M T X solution is pH dependent and subject to general acid-base catalysis by most of the buffer substances. Acetate, phosphate, and borate catalysis increases linearly with increasing catalyst concentration, while a non-linear variation of degradation rate with increasing carbonate buffer concentration has been observed (Hansen et al., 1983).  The maximum stability of M T X is between pH 6.6-8.2 (Hansen,  et al., 1983). For an isotonic buffer-free M T X solution at pH 8.5 and 25°C, the shelf-life for 10% degradation of the M T X is estimated to be 54 months.  At pH above 6.5,  N -methyl-folic acid is the only degradation production while in acidic solutions several 10  compounds are formed (Hansen et ah, 1983). M T X undergoes photodegradation when stored in diluted solutions under light (Chatterji and Gallelli, 1978).  Photodegradation can take place under normal lighting,  but is more rapid in direct sunlight, with about 11% drug loss from a 1 mg/mL solution after 7 h.  Storage under normal lighting resulted in little change in drug concentration  over 24 h and a decrease of up to 12% by 48 h (McElnay, 1988).  The major degradation  products following photolysis are p-aminobenzoylglutamic acid and 2,4-diamino-6-pteridincarbaldehyde (Dyvik, etal., 1986).  22  A) Methotrexate pteridine ring  r  p-aminobenzoic acid  i  NH, -C  H,N2  /  glutamic acid  r  r  -N  \  X  C- —CH 9  2  N/,o  =CH 7  \  H,C  /  CH=CH \  /  ~\ //  C H 2 CH2  \  / NH-HC  CH—CH  \  / B)  -OH  _  -OH  folic acid  C) folic acid (reduced form)  OH N  C  C' 2  \  N  N=C ,  C  N  N  /  5  \ N  \  C  CH=CH / \ ~ C\ /C  / 6  N=CH  \  7  //  //  C  \  CH—CH  D)  ^  O  OH  CH? CH? -  /  /  C  H N2  C' 2  \  /  C  X  C - -N \  /  5  9 \  -CH, 7  NH-HC\  \\ C CH=CH O / / \ // NCH -CH / 1 0 -C\ ,C—c / NH-HC  CH—CH  / .  '  2  2  2  \  -OH  //  CH—CH  \  \  7-OH-methotrexate  J  2  N- - C  / Vv  \ _ /  c  .  CH, \ N H,C  CH=CH / \ C> ,CCH—CH  V  // •~C  \  CH? CH? —  /  NH-HC \  0t  Figure 1.5  -OH  / -OH  Chemical structures of A) methotrexate, B) folic acid, C) folic acid (reduced  form), D) 7-hydroxy methotrexate (7-OH-MTX).  23  2  -OH  OH  1.4.2  Formulations  M T X is currently formulated as tablets for oral administration at strengths of 5, 7.5, 10 and 15 mg. Methotrexate Sodium Injection is formulated as 25mg/mL and lOmg/mL solutions. Nonmedicinal ingredients include sodium chloride and water for injection. Sodium hydroxide and/or hydrochloric acid maybe used for pH adjustment.  Benzyl  alcohol may be used as a preservative.  1.4.3  Pharmacology and indications  M T X is a useful drug in the management of acute lymphoblastic leukemia in children and in choriocarcinoma and related trophoblastic tumors of women.  Beneficial  effects also are observed in patients with osteosarcoma and mycosis fungoides and when M T X is used as part of the combination therapy in Burkitt's and other non-Hodgkin's lymphomas and carcinomas of the breast, head and neck, ovary, and bladder (Chabner et al., 1996). M T X has also been used in the treatment of severe, disabling psoriasis (Chabner et al., 1996). The use of M T X in the treatment of rheumatoid arthritis was approved by the Food and Drug Administration in 1988 (van Ede et al, 1998). The dose of M T X in the treatment of psoriasis in adults is 2.5 mg orally for 5 days following a rest of period of at least 2 weeks (Chabner et al., 1996). The use of M T X in the treatment of rheumatoid arthritis in adults is 10 to 15 mg/week and the maximum  24  dose is 25 to 30 mg/week (van Ede et al, 1998). The doses used for the treatment of cancers are much higher.  For example, in the combination therapy of leukemias and  non-Hodgkin's lymphomas, 6 to 72-hour infusions of 250 mg to 7.5 g/m of M T X may 2  be employed (Chabner et al, 1996). M T X is a folic acid (shown in Figure 1.5 B) analog in which the groups bonded to the C4 carbon and N i n hydrogen are N H 2 and C H 3 , respectively.  Reduced folic acids  (tetrahydrofolates) are essential cofactors that provide single carbon groups in several reactions involved in the synthetic pathways for purines, pyrimidines, formation of polyamines, and transmethylation of phospholipids and proteins (Hillson and Furst, 1997). The enzyme dihydrofolate reductase is responsible for reducing partially oxidized folates (dihydrofolates) to tetrahydrofolates.  M T X inhibits dihydrofolate  reductase with high affinity, resulting in depletion of tetrahydrofolates that are required for the synthesis of purines and thymidylate (van Ede et al., 1998). A portion of intracelluar M T X is metabolized to polyglutamates and these polyglutamates of M T X also inhibit the folate-dependent enzymes of purine and thymidylate metabolism. Subsequently, the synthesis of D N A and RNA, as well as other vital metabolic reactions are interrupted (Chabner et al, 1996; Cronstein, 1997; Hillson and Furst, 1997).  25  When M T X was first introduced in the treatment of rheumatoid arthritis in the 1980's, the rationale was that it inhibited proliferation of the lymphocytes or other cells responsible for inflammation in the joint.  However, over the past 15 years, several lines  of evidence have clearly suggested that M T X does not act simply as an antiproliferative agent for the cells responsible for the joint inflammation in rheumatoid arthritis (Cronstein and Merrill, 1996; Cronstein, 1997). The rapid clinical remission and the short term effect on the acute phase reactants, as seen with low dose M T X administration in most patients with rheumatoid arthritis, as well as the rapid flare of disease after drug discontinuation, suggest that the mechanism of the action of low dose M T X might be more anti-inflammatory than antiproliferative (Kremer, 1994; Cutolo et al., 2001). Although the exact mechanisms are still not fully understood, the anti-inflammatory effects of M T X are speculated to be related to adenosine induced immunosuppression. M T X inhibits both dihydrofolate reductase and other folate dependent enzymes and leads to adenosine overproduction (Furst, 1997; Hillson and Furst, 1997; Cutolo et al., 2000). Extracellular adenosine can bind to adenosine surface receptors types A l , A 2 a , A2(3, A3, which have been found on many different cell types such as fibroblasts and endothelial cells (van Ede et al., 1998). Low dose M T X exerts its anti-inflammatory effect by inducing extracellular adenosine, which acts predominantly through A 2 receptors and  26  inhibits phagocytosis and secretion of TNF-cc, IFN-y, IL-2, IL-12 and many, other proinflammatory cytokines (Cutolo et al, 2001).  Through adenosine mediated  pathways, M T X treatment in rheumatoid arthritis seems to decrease the monocytic and macrophagic cytokines IL-1, IL-6 and T N F a secretion and increase IL-1 receptor antagonist production (Cronstein, 1997).  At the same time, M T X increases  anti-inflammatory IL-4 and IL-10 gene expressions and decreases gene expression of proinflammatory T h l cytokines such as IL-2 and IFN-y, with resulting anti-inflammatory effects (Seitz et al, 1995; Seitz et al, 1996; Cutolo et al, 2001). As for antiproliferative effects, intermediate M T X concentrations (50 pg/mL), as obtained in serum after low dose treatment, can induce both a significant cell growth inhibition and apoptosis in immature T h l monocytic cells but has no significant effect on synovial macrophage proliferation (Cutolo et al, 2000).  This finding suggests that  M T X might inhibit recruitment of immature and inflammatory monocytes into inflammatory sites and could reduce the survival of these cells in the inflamed synovial tissue (Cutolo et al, 2000). Low dose M T X in rheumatoid arthritis treatment seems to exert anti-inflammatory effects by acting at different levels of the pathophysiological cascade. The direct inhibitory effects on proliferation and the induction of apoptosis in cells involved in the  27  immune/inflammatory reaction represent the first step of the intervention while the inhibition of both monocytic /lymphocytic proinflammatory cytokines involved in rheumatoid synovitis, seems to be the key role in the sustained anti-inflammatory actions exerted by low dose M T X (Cutolo et al, 2001).  1.4.4  Toxicity  When used in low doses for the treatment of rheumatoid arthritis, M T X has proved to be a very effective, fast working disease modifying agent with an excellent efficacy-toxicity ratio. Nevertheless, the main reason for discontinuation of M T X is not the lack of efficacy, but toxicity.  In approximately 30% of rheumatoid arthritis patients,  toxicity leads to discontinuation of M T X therapy (van Ede et al, 1998). Table 1.1 outlines the range of toxicities of methotrexate found in the medical literature (Furst, 1997).  Gastrointestinal toxicities including stomatitis, nausea and  abdominal distress are most common. Although the precise mechanisms of toxicity are still not clear, some side effects have been directly related to M T X involvement in the previously mentioned metabolic pathways.  Side effects of M T X such as GI distress and  bone marrow suppression seem to be directly related to folate antagonism and its cytotoxic effects, especially in tissues with a high cell turnover such as bone marrow and the gastrointestinal tract, which have a high requirement for purines, thymidine and  28  pyrimidine (O'Dell, 1997; van Ede et al, 1998). Adenosine release in the central nervous system and its activation of A l receptors in the brain may be responsible for induction of fatigue and lethargy (Grim et al, 2003). A l receptors are also present in endothelial cells and their activation provokes vasodilatation and therefore explains the headache that appears in many patients a few hours after administration of low dose M T X (Grim et al, 2003). Other side effects like pneumonitis and progressive subcutaneous nodules probably have a more complex origin (O'Dell, 1997; Salaffi et al, 1997). M T X can induce acute parenchymal damage and fibrosis in the liver.  In a  clinical setting, serum transaminases, aspartamine transferase (AST) and alanine transferase (ALT) are often monitored (Kirchain and Gill, 1996).  29  Table 1.1  Adverse effects when using methotrexate for the treatment of rheumatoid  arthritis (Furst, 1997). Methotrexate dose (mg/week) 7.5-48 Duration of use (months) 4-183  % of incidence  Adverse events Central nervous system  13-35  (headache, fatigue, "fuzziness", malaise) Gastrointestinal (GI)  19-65  Nausea, GI distress  19-65  Stomatitis  2-55  Hematological Anaemia  1 -2  Leucopenia  2-21  Thrombocytopenia  1-5  Infection  Rare  Integument Alopecia  1-6  Rash  2-15  Liver Cirrhosis  Rare  Elevated transaminases  8-30  Osteopathy  Rare  Pulmonary  Rare  Pneumonitis (hypersensitivity)  1 -7  Pseudolymphoma  Rare  Teratogenicity  Definite  30  1.4.5  Pharmacokinetics  In the treatment of rheumatoid arthritis, M T X is usually given orally. In adult patients, after oral administration, active absorption of the drug occurred in the proximal jejunum (Hillson and Furst, 1997; Grim et al, 2003).  When given intramuscularly or  subcutaneously, the drug was absorbed more rapidly and reached higher serum concentrations compared with the oral route (Grim et al, 2003).  Nevertheless, the  mean absolute bioavailability was about 70-80% regardless of the route of administration and a large inter- individual variation between 30-90% has been observed (Grim et al., 2003). In blood, 30-70% of M T X was bound to proteins, almost exclusively to albumin (Hillson and Furst, 1997; Grim et al, 2003).  The volume of distribution at steady state  ranged from 0.33-1 L/kg in 10 patients receiving a single 7.5 mg dose of M T X (Sinnett et al, 1989). Enterohepatic cycling of the drug has been suggested since a significant increase in drug plasma concentration was observed 8 h following M T X administration (Edno et al, 1996).  The concentrations of M T X in the synovial fluid were  approximately equal to plasma concentrations at 4 and 24 h after oral or intramuscular administration (Herman et al, 1989).  31  M T X is transported into cells by passive transmembrane diffusion and by a folate surface receptor mediated active transport system (Bannwarth et al, 1996). Once inside the cell, up to six glutamate residues may be added to the drug molecule. M T X polyglutamates cannot be transported extracellularly unless they are hydrolysed back to the monoglutamate (Hillson and Furst, 1997; Grim et al, 2003). M T X has been found to accumulate in erythrocytes (Kremer et al., 1986), peripheral blood T-lymphocytes, fibroblasts, myeloid precursors in bone marrow and keratinocytes (Grim, et al., 2003). Three metabolic pathways for M T X have been described in humans.  First, the  drug may be metabolized by intestinal bacteria to 4-amino-4-dexoy-N -methyl-pteroic 10  acid and this metabolite usually accounts for less than 5% of administrated dose (Grim et al., 2003).  Secondly, 3-11% of a M T X dose is converted by an aldehyde oxidase in the  liver to 7-hydroxy-methotrexate (7-OH-MTX) (Figure 1.5) (Hillson and Furst, 1997). 7-OH-MTX is poorly water soluble and a 10 fold less potent inhibitor of dihydrofolate reductase as compared to M T X (Furst, 1997). Due to its slower rate of urinary excretion, plasma concentrations of 7-OH-MTX usually exceed those of M T X within 8-10 h after drug administration and thus 7-OH-MTX is the major circulating metabolite of M T X (Bannwarth et al, 1996; Grim et al, 2003). Thirdly, M T X is converted to  32  polyglutamates intracellularly. M T X polyglutamates have 7 fold higher inhibitory effects for dihydrofolate reductase than M T X (Furst, 1990). Elimination of M T X from plasma was shown to be biphasic or triphasic depending on the length of sample collection period. The elimination half-lives of the (3 phase (ti/ (3 ) for M T X and its 7-OH-MTX were 5 to 8 h and 8 to 11 h, respectively in 2  rheumatoid arthritis patients (Edno et al., 1996; Grim et al., 2003). sampling, the longer the reported  ti/2P  The longer the  of the drug, probably due to intracellular M T X  storage, polyglutamylation and slow release back to plasma (Shen and Azarnoff, 1978; Grim et al., 2003). 2003).  Renal excretion is the major elimination route for M T X (Grim et al.,  This route of elimination accounts for 60-90% of the M T X excretion, while  biliary elimination is responsible for up to 10 to 30% of M T X excretion (Bannwarth et al., 1996). In the kidneys, M T X is filtered in the glomeruli and undergoes active secretion and active reabsorption (Edno et al., 1996). Both secretion and reabsorption in renal tubules can be saturated even at low M T X plasma concentrations within the range of 0.1-1 umole/L. Thus non-linear elimination may result following the administration of 7.5-30 mg of M T X and contribute to the interindividual variability in M T X concentrations (Hendel andNyfors, 1984).  33  1.5 1.5.1  INTRA-ARTICULAR DRUG DELIVERY Synovial permeability and effects of inflammation  The synovium has a complex physiology and can be likened to a double barrier consisting of a capillary endothelium (first barrier), which controls the transit of large molecules such as proteins, and the synovial interstitium (second barrier) which limits the transit of small hydrophilic solutes (Netter et al, 1989). Synovial inflammation induces anatomical changes such as synovial cell proliferation, cellular infiltration, villous hypertrophy, and angiogenesis (Smolen and Steiner, 2003). Levick (1995) showed that in normal human synovium, the capillaries are close to the joint cavity with a modal depth of 36 pm, while in chronic rheumatoid arthritis, the capillaries become buried under the hypertrophied synovial lining and the mean distance from the surface is more than twice as great.  Stevens et al (1991) showed  that the synovial thickening and cellular infiltration in chronic rheumatoid arthritis caused a greater than two fold increase in average transport distance to the joint cavity and a reduction in capillary density to one-third of normal. Simkin and Pizzorno (1974) have shown that most small molecules cross the synovium in both directions by passive diffusion, and are limited primarily by the relatively long, narrow diffusion path between synovial lining cells and not by any part of  34  the microvascular barrier.  Comparing synovial permeabilities in normal subjects and  patients with rheumatoid arthritis, Simkin et al (1979) have shown that the synovium in joints of arthritic patients was less permeable to small molecules such as tritiated water, glucose or urate than that of normal individuals.  On the other hand, total proteins  moved across arthritic synovium faster than normal synovium. It was suggested that small molecules left the rheumatoid joint space slowly because the synovial proliferation characteristics of the disease decreased the diffusion path between synovial cells. The permeability to proteins was increased due to increased protein permeability of the microvascular barrier (Simkin, 1979). Vascular wall changes secondary to disease did not increase the delivery of small molecules because the vessel wall is generally not a significant barrier to small molecules (Simkin, 1979; Simkin and Nilson, 1981).  1.5.2  Intra-articular drug therapy  Intra-articular therapy has been used for patients in whom the arthritis manifests in only a limited number of joints.  The rationale for intra-articular therapy is to target the  drug to the site of action and to minimize systemic toxic effects of the drug (Itoh, 1992). The intra-articular injection of corticosteroids has played an important role in the management of inflammatory arthritis at an early stage. Injections of corticosteroids into joints have been shown to be effective in easing the pain and stiffness and may even  35  improve advanced destructive arthritis (Derendorf et al., 1986).  Microcrystalline  steroids such as triamcinolone, methylprednisolone, and rimexolone are commonly injected intra-articularly as suspensions of poorly water soluble crystals, and the duration of benefits can be up to several months depending on the pharmacokinetic properties of the particular steroid (Hunter and Blyth, 1999). On the other hand, the response to a soluble hydrocortisone succinate injection was very brief due to the rapid clearance of hydrocortisone solution from the joint cavity (Hunter and Blyth, 1999). Pharmacokinetic studies of intra-articular steroid suspensions have shown that the rate-limiting step for steroid disposition in the joint was the dissolution rate in the synovial joint (Kahn et al., 1970; Derendorf et al., 1986). In the plasma concentration-time profiles, the terminal part of the curve represented systemic absorption from the joint rather than elimination, while the early phase of the curve represented the elimination ("flip-flop-case") (Derendorf et al, 1990). Despite their effectiveness in suppressing pain, swelling and redness, intra-articular steroids produced an acute flare following injection due to their crystalline nature (Kahn et al., 1970). They have also been shown to produce damage to the articular cartilage and increase the risk of infection in the joint (Hunter and Blyth, 1999).  36  Intra-articular injection of hyaluronic acid has been approved as a viscosupplement in the management of osteoarthritis (Gossec and Dougados, 2004). Hyaluronic acid may help restore the viscoelasticity of the synovial fluid and promote the endogenous synthesis of a high molecular weight, and possibly more functional hyaluronan, thereby improving mobility and articular function and reducing pain (Gossec and Dougados, 2004).  Studies have been conducted to investigate the efficacy of hyaluronic acid in  controlling the symptoms of rheumatoid arthritis (Goto et al., 2001; Tanaka et al., 2002). The results showed that following 5 consecutive injections of 25 mg per week of high molecular weight sodium hyaluronate, patients had significant improvement in pain symptoms, and inflammation. The elimination of radio-labeled hyaluronic acid from the knee joints of healthy men has been demonstrated to be triphasic with half-lives of 1.5 h, 1.5 days and 4 weeks.  A n increase in intra-articular fluid volume at 24 h post  injection was also observed (Lindqvist et al, 2002). M T X has been administered as an intra-articular injection but the results have not been very encouraging (Marks et al, 1976; Bird et al, 1977). Intra-articular injection of 5 mg M T X solution and intra-articular 20 mg triamcinolone hexacetonide were compared in arthritic patients with persistent bilateral knee effusions and showed that M T X had no immediate anti-inflammatory effect, nor did it give the relief of intra-articular steroid  37  (Bird et al., 1977). Intra-articular combination therapy of 20 mg triamcinolone and 50 mg M T X in arthritic patients reduced joint effusion and knee circumference compared to triamcinolone alone (Blyth et al., 1998). Wigginton and coworkers (1980) studied M T X pharmacokinetics after intra-articular injections in patients with rheumatoid arthritis.  Patients were  administered 5 mg of M T X every 24 h into the same knee joint up to 48 h. The synovial fluid and serum M T X concentration profiles showed that the elimination of M T X from the joint was biphasic over 24 h, with half-lives of 0.54 and 2.90 h. The intra-articular apparent volume of distribution was calculated to be 69 mL and the intra-articular M T X clearance was 0.28 mL/min.  The authors concluded that intra-articular M T X was  clinically ineffective, primarily because the intra-articular half-life of M T X was too short relative to the probable synovial cell cycle generation time (Wigginton et al., 1980). Repeated intra-articular M T X doses have produced better results (Gao et al., 1998; Iagnocco et al., 2005). Rheumatoid arthritis patients were treated with up to 6 intra-articular injections of 10 mg M T X every 3 to 7 days and their synovial fluids were analysed for leukocyte counts and cytokine levels (Gao et al., 1998). The granulocyte counts and IL-8 levels decreased in all M T X treated patients, whereas IL-6 and IL-10 showed only minor changes.  IL-8 has been reported to be related to the activity of the  38  inflammatory process (Verburgh et ah, 1993).  Iagnocco and coworkers (2005) have  demonstrated that multiple intra-articular injections of 10 mg M T X every week for 8 weeks in rheumatoid arthritis and psoriatic arthritis patients resulted in a decrease in local and systemic inflammation. There was a significant reduction in synovial thickness and joint effusion and an improvement in maximum knee joint flexion following treatment (Iagnocco et al., 2005). Liposomes encapsulating M T X have been formulated for intra-articular delivery (Foong and Green, 1993; Williams et al, 1996). Increased retention of M T X in inflamed rabbit knee joints was observed following intra-articular injection of liposomal M T X (Foong and Green, 1988). This formulation reduced joint swelling, synovial fluid production, and synovium proliferation after intra-articular injection in rabbits with established antigen-induced arthritis (Foong and Green, 1993). Williams et al (1996) showed that a single intra-articular injection of M T X encapsulated in multilamellar liposomes was retained within the joint and produced a rapid reduction in knee swelling of rats within 24 hours, and a progressive reduction in joint swelling over the next 20 days.  39  1.6 POLYMERIC DRUG DELIVERY SYSTEMS 1.6.1  Controlled release drug delivery systems  Controlled release drug delivery occurs when a polymer, whether natural or synthetic, is combined with a drug or other active agent in such a way that the active agent is released from the material in a designed manner (Robinson, 1997).  Goals of  controlled release drug delivery systems include, maintaining the drug in the desired therapeutic range with just a single dose, localizing delivery of the drug to a particular body compartment, reducing the need for follow-up care, preserving medications that are rapidly destroyed by the body, and increasing patient comfort and improving compliance (Langer, 1990). The ideal drug delivery system should be inert, biocompatible, possess appropriate mechanical properties, comfortable for the patient, capable of achieving high drug loading, safe from accidental drug release, simple to administer and remove, and easy to fabricate and sterilize (Robinson, 1997).  1.6.2  Biodegradable polymers for drug delivery  A range of polymers have been employed to control the release of drugs and other active agents. To be successfully used in controlled drug delivery formulations, the polymer must be chemically inert and free of leachable impurities (Piskin, 1994). It  40  must also have an appropriate physical structure, with minimal undesired aging, and be readily processable.  More and more polymers are designed to degrade within the body  so that they do not require retrieving following administration (Robinson, 1997). Most commonly used biodegradable polymers include poly(lactic acid)(PLA), poly(glycolic acid) (PGA) and the copolymers of P L A and PGA, poly(lactic-co-glycolic acid) (PLGA) (Figure 1.6).  41  B)  A)  O  O -C  -C—C  CH—O-  o  O  * -CH-  O  H  I  ,1  -C-  -C-  -O-  i  H  m  CH,  Figure 1.6  CH  H  CH,  -C-  H  The chemical structures of A) poly(lactic acid) (PLA), B) poly( glycolic  acid) (PGA), and C) poly(lactic-co-glycolic acid) (PLGA).  42  * indicates the chiral center.  1.6.2.1 Structure and molecular weight ofpolymers Polymers are large molecules built up by the linking together of large numbers of much smaller molecules, which are termed monomers (Odian, 1991). the polymer depends on the monomers used in its preparation. have hundreds or more repeating units.  The structure of  Typically, polymers  If the polymer is prepared from a single  monomer the product is referred to as a homopolymer.  For example, P L A and P G A are  both homopolymers with the repeating units being lactic acid and glycolic acid, respectively (Figure 1.6).  If more than one monomer is employed, the product is a  copolymer such as P L G A (Figure 1.6). With only a few exceptions, a synthetic polymer is always a mixture of molecules with different molecular weights in different amounts and distributions.  Therefore,  there is no well-defined molecular weight and only averaged values can be given (Allcock and Lampe, 1981).  The processing behavior and many end-use properties of  polymers are influenced not only by the average molecular weight but also by the molecular weight distribution. Some properties, including tensile and impact strength, are strongly influenced by the short molecules; for other properties, such as solution viscosity and low shear melt flow, the influence of the middle length of the chains is  43  predominant; yet other properties, such as melt elasticity, are highly dependent on the amount of the longest chains present (van Krevelen, 1997). Several characteristic molecular weight averages exist.  Most important are the  number and the weight averages of the molecular weight, M n and M w , respectively. The number average molecular weight (Mn) is defined as the total weight of all the molecules in a polymer sample divided by the total number of moles present, as follows: M n = W/N = S.niMi/ Zni  Equation 1.1  Where W is the total weight of polymer sample, N is the total number of moles and ni is the number of moles of molecular weight M i . Weight average molecular weight, (Mw) is defined as M w = SwiMi/W = S n i M i / SniMi 2  Equation 1.2  Where wi is the weight of molecules of molecular weight M i . M w is an average which is weighted so that the contribution of each chain length depends on its proportion by weight in the total sample.  Thus, a small chains would  have the same contribution as one chain ten times its length. The molecular weight of a polymer is also expressed by its intrinsic viscosity [r\]. For a linear, unbranched, and non cross-linked polymer of the same chemical constitution, the viscosity of a dilute polymer solution is directly related to the molecular  44  weight average.  By determining the relative viscosity using a viscometer, several  viscosity numbers can be calculated (Allcok and Lampe 1981). The relative viscosity, r| i is the ratio of the viscosity of a diluted polymer solution (r\) to the viscosity of the re  pure solvent (r|o) as shown in Equation 1.3.  The specific viscosity, r| , is the fractional sp  increase in viscosity caused by the presence of the dissolved polymer in the solvent as shown in Equation 1.4.  The specific viscosity and the relative viscosity clearly depend  on the concentration of the polymer in solution.  The quantity r | / C , where C is the sp  concentration of polymer in g/cm , is sometimes called the reduced viscosity. The 3  intrinsic viscosity, [r|], is defined as the limit of the reduced viscosity as the concentration approaches zero (Equation 1.5) (Allcock and Lampe, 1981). r] = —  Equation 1.3  rel  ^ = ^  " 7o  2  =^ - l  Equation 1.4  Equation 1.5  Two methods have been used to determine [r|]: the extrapolation method and the single point method.  The extrapolation method involves measuring r| i, for a polymer re  solution at a series of concentrations (Collins et al., 1973). The single point method uses Equation 1.6 as shown below. This equation is only valid for solutions of polymer in a  45  good solvent, and the concentration should be chosen such that r\ < 0.2 (Solomon and sp  Ciuta, 1962).  Equation 1.6 c  Intrinsic viscosity can be used to calculate a molecular weight average using the Mark-Houwink equation: [rj]  = K M v 'a  Equation 1.7  Where K and a are constants and M v is the viscosity average molecular weight.  The  constants K and a are specific for a given polymer/solvent system at a particular temperature. If these values can be obtained from the literature, then M v can be calculated (Allcock and Lampe, 1981). Gel permeation chromatography (GPC) is a process for the separation of polymers according to their molecular size. Using this method, the rate of permeation of polymer chains in solution through a gel packed with microporous beads of fixed particle and pore sizes, expressed as retention time, is related to the hydrodynamic volume of each chain. The time for passage of the polymer molecules through the gel column decreases with increasing molecular weight.  A calibration curve for a given column may be  constructed by measuring the retention time of a series of monodisperse polymer standards of known molecular weight (MW) and plotting log(MW) versus retention time. 46  After calibrating the column, the molecular weight (both M n and Mw) and the polydispersity of polymer samples can be determined (Collins et al., 1973). 1.6.2.2  Polymer morph ology  Polymer morphology describes the arrangement in three dimensions of polymer chains with respect to long range order (Rosen, 1993).  Most polymers show,  simultaneously, the characteristics of both crystalline solids and highly viscous liquids. The known polymers range from those that are completely amorphous, to semicrystalline polymers with crystallinities from low to high. Two models of crystallinity have been used to describe the nature of semicrystalline polymers (Rosen, 1993).  The "fringed-micelle model" assumes that polymers consist of  small-sized, ordered, crystalline regions embedded in a disordered, amorphous polymer matrix.  Crystallites are formed when chain segments from different polymer chains are  precisely aligned together and undergo crystallization. Each chain can pass through several different crystalline regions and contribute ordered segments to several crystallites.  The segments of the chain in the regions between the crystallites make up  the disordered amorphous matrix (van Krevelen, 1997) (Figure 1.7A). In the "chain folded model" of polymer crystallinity, a single polymer chain folds upon itself to form lamellar crystallites interspersed in an amorphous phase.  47  Several chains are involved in  each crystallite, and extend into the amorphous regions of the matrix (Rosen, 1993) (Figure 1.7B).  48  A)  lOUS  Crystalline region  B) O O  /"> o O O O O /"\ ' o- O O O O O O '  oin/ninnini/-ii/^n * o o o o o o o o o o o o o o o o o o o o O o o o o o o o o o o 0 0 0 0 0 0 0 0 0 0 0  > ** . jr _J> *  Chain folded lamellar crystallite  Figure 1.7  Schematic representation of A) fringed micelle model B) the chain folded  model of polymer crystallinity.  49  1.6.2.3  Thermal transitions  Polymeric materials are characterized by two major types of thermal transitions, glass transitions and crystalline melting transitions.  Whether a polymer sample exhibits  both transitions, or only one, depends on its morphology. polymers show only a glass transition. show only a melting transition.  Completely amorphous  A theoretically 100% crystalline polymer would  Semicrystalline polymers exhibit both the crystalline  melting and glass transitions. In an amorphous polymer, at temperatures below the glass transition temperature (Tg), the polymer chain segments are frozen in fixed positions.  Some molecular  movements of chain segments take place in the form of vibration about a fixed position. The polymer is rigid and glassy. segmental vibrations increases.  With an increase in temperature, the amplitude of At the glass transition temperature, the chain segments  have sufficient energy to overcome the secondary intermolecular bonding forces resulting in a transition from the glass to the rubber stage. In the rubbery state, the segmental motions are very rapid whereas the motion of the entire molecule is restricted by chain entanglement (Allcock and Lampe, 1981). Many physical properties change profoundly at the glass transition temperature, including heat capacity, refractive index, mechanical damping, and electrical properties  50  (Nicholson, 1997). A l l of these are dependent on the relative degree of freedom for molecular motion within a given polymeric material and each can be used to monitor the point at which the glass transition occurs. The melting of a semicrystalline polymer takes place over a wider temperature range than that observed for small organic compounds, due primarily to the presence of crystallites of different sizes and the more complex process for melting of large molecules (Allcock and Lampe, 1981). The crystalline melting temperature, Tm, is generally reported as the temperature of the onset of melting. structure on Tm is similar to that on Tg. Tm values.  The effect of polymer  Polymers with low Tg values usually have low  The two thermal transitions are generally affected in the same manner by  the molecular weight, the molecular symmetry, structural rigidity, and secondary bonding forces of polymer chains.  For linear polymers, the Tg and Tm values are increasing  functions of the molecular weight (Billmayer, 1984; Mark et al, 1984). Among commonly used synthetic biodegradable polymers, poly(glycolic acid) (PGA) is a semicrystalline polymer with a Tm of 224-226°C and a Tg of 36°C (Piskin, 1994).  Poly(lactic acid) (PLA) is a chiral molecule and can exist as two optically  active stereoregular polymers, poly(D-lactic acid) and poly(L-lactic acid) (PLLA) and the racemic form, poly(D,L-lactic acid). For biomedical applications, P L L A and poly  51  (D,L-lactic acid) have been the most widely studied (Piskin, 1994). P L L A polymers are semicrystalline polymers with a Tm of 184°C and a Tg of 57-60°C.  Poly(D,L-lactic acid)  is amorphous and has a Tg of 54-59°C (Holland and Tighe, 1992; Piskin, 1994).  For  P L G A , the copolymers of lactic acid and glycolic acid, the degree of crystallinity and Tg varies with the ratio of glycolic acid and lactic acid (Holland and Tighe, 1992; Piskin, 1994). 1.6.2.4  Biodegradation  Biodegradation is a broad term that refers to hydrolytic, enzymatic or bacteriological degradation processes occurring in a polymer matrix (Holland and Tighe, 1992). The most common degradation mechanism for biodegradable polymeric drug delivery systems is hydrolytic degradation.  Hydrolytic biodegradation is dependent on  the ability of water to insert itself into a susceptible functional group in the polymer backbone, which results in cleavage of the bond. Factors affecting the rate of hydrolysis of the functional group include hydrophobicity of the polymer, steric hindrance of the polymer side chains near the functional group, the crystallinity of the polymer and environmental factors such as temperature, pH and ionic strength (Kakino et al, 1986; Holland and Tighe, 1992; Pistner et al, 1993).  52  For P L A , hydrolytic degradation is considered to be the major biodegradation mechanism.  The cleavage of an ester bond yields carboxyl and hydroxyl end groups.  The carboxyl end groups are capable of catalyzing hydrolysis of other ester bonds, a phenomenon called autocatalysis (Li, 1999). The hydrolytic degradation starts with a random chain scission process as the ester bonds are hydrolyzed by the water molecules. The molecular weight of the polymer decreases significantly, but there is initially no appreciable weight loss and no soluble monomer products are formed.  When the  molecular weight drops sufficiently, the degradation products become soluble or separate from the matrix, and loss of material will occur.  Soluble monomer products are then  formed from soluble oligomeric fragments (Jalil and Nixon, 1990).  Loss of mass from  an implanted material is called bioerosion. Bioerosion may occur as a result of degradation but it may also occur with non-degradable materials, which dissolve or disintegrate in the body (Jalil and Nixon, 1990). The final elimination phase in which dissolved monomers and oligomers of polymer are absorbed is called bioresorption (Vert et al, 1992).  Small particles of P L A and  dissolved oligomers are taken up into macrophages for subsequent digestion by lysozomal enzymes (Woodward et al., 1985; Bos et al, 1991).  The final degradation  product of P L A is lactic acid, which enters the tricarboxylic acid cycle and is metabolized  53  and subsequently eliminated from the body as carbon dioxide and water (Jain et al., 1998). The effects of biodegradation and subsequent bioerosion of biodegradable drug delivery systems are seen as changes in inherent viscosity and molecular weight of the polymer and in mechanical properties such as tensile strength and drug release kinetics (Holland and Tighe, 1992). 1.6.2.5  Biocompatibility  A l l biomaterials are evaluated in terms of biocompatibility depending on the intended medical application (Park and Park, 1995). Biocompatibility can be defined as the acceptance of an artificial material by the surrounding tissues and by the body as a whole (Wang et al., 2004). The term biocompatibility encompasses many different properties of the materials, including toxicity, blood compatibility, and tissue compatibility (Wang et al., 2004). When a biomaterial is exposed to blood, certain blood proteins adsorb rapidly to the biomaterial, and protein adsorption, depending on the type of adsorbed proteins, is followed by platelet adhesion.  The activation of adherent platelets leads to the  formation of thrombi on the surface, which may result in blockage of blood supply or may affect the drug release profiles of the drug delivery systems (Park and Park, 1995).  54  Tissue damage created by the implantation procedure usually results in inflammation (Park and Park, 1995). The inflammatory process is accompanied by a series of defensive reactions by neutrophils, eosinophils, macrophages and foreign body giant cells.  Macrophages initiate the repair of damaged tissue by forming the scaffold  for repair, which is called granulation tissue.  If the implant is not phagocytosed by the  cells, the body tends to completely isolate the foreign implant by forming a sheath-like fibrous membrane capsule around the implants, which is termed scar tissue (Park and Park, 1995).  The fibrous capsule often contracts and causes pain in patients and  deformation of the implant.  For a drug delivery implant, the fibrous capsule may alter  the drug release kinetics (Park and Park, 1995). The biocompatibility of P L A implants was first noted by Kulkarni and coworkers (1971).  In vivo evaluation of P L A drug delivery systems, such as microspheres, was  carried out by several groups of workers (Ratcliffe et al., 1984; Jain, 2000; Johansen et al., 2000; Liggins et al., 2004).  In general, tissue responses were mild with no abnormal  reactions leading to the rapid clinical acceptance of lactic acid polymers. In vivo evaluation of P L L A bone fixation plates and screws three years after implantation, showed a foreign body reaction at the site of the implant although the presence of inflammation was considered minimal (Bergsma et al, 1993).  55  P L L A microspheres in  the size range of 1-80 um injected subcutaneously in mice, caused a mild inflammatory response initially and the response disappeared 6 months following injection. Histology showed a fine capsule around the implant at all time points but no remnants of scar tissues were discovered, illustrating excellent biocompatibility of P L L A (Lemperle et al., 2004). Hooper et al (1998) investigated the biocompatibility of P L L A implants as a function of molecular weight of P L L A .  The results showed that the tissue response to  the implant fluctuated as a function of the degree of degradation, exhibiting an increase in the intensity of inflammation as the implant began to lose mass. A thick capsule containing fibroblasts and macrophages was observed surrounding the implant.  There  was no ingrowth of tissue into the implant.  1.6.3  Mechanisms of drug release from polymeric drug delivery systems  Polymeric controlled release drug delivery systems can be generally divided into two types; reservoir-type and matrix type.  In the reservoir-type system, the drug is  encapsulated inside a system enclosed by a polymeric membrane and the rate of drug release is controlled by its permeation through the membrane wall.  In the matrix-type  drug delivery system, the active agent is homogeneously dispersed throughout a polymeric matrix (Chien, 1982).  In this thesis, the discussion is focused on the  matrix-type drug delivery system.  56  Three different mechanisms of drug release can be identified and are referred to as diffusion controlled, chemical controlled and solvent controlled release.  These  classifications represent theoretical situations where the rate of drug release is controlled predominantly by the diffusion of drug through a polymeric matrix or a membrane, the chemical processes such as polymer degradation, erosion or the cleavage of a drug from a polymeric carrier, and solvent interactions such as swelling of the polymer, respectively (Sinkon and Kohn, 1993). 1.6.3.1  Diffusion controlled drug release  In the diffusion controlled system, if the drug to be released is dispersed uniformly throughout the polymer, the system is called a monolithic diffusion system (Baker, 1987). A monolithic system can be either a monolithic solution in which the drug is dissolved in the polymer or a monolithic dispersion in which only a portion of the drug is dissolved in the matrix and the remainder is dispersed as particles throughout the matrix (Chien, 1982; Baker, 1987). The release of solutes from matrix systems is based upon Fick's laws of diffusion.  Fick's first law states that the flux (J) or the rate of solute transfer across a  plane of unit area is given by: J - -D^dx  Equation 1.8  57  where dC/dx is the concentration gradient or the change in concentration (C) with respect to distance (x), and D is the diffusion coefficient.  In the case of a monolithic solution,  no concentration gradient exists in the matrix prior to the onset of drug release.  As the  drug begins to be released from the surface of the matrix, a concentration gradient is established.  Drug begins to diffuse down the concentration gradient from the interior of  the matrix towards the surface and is gradually released at the surface (Baker, 1987). In monolithic dispersion systems, the amount of initial drug loading has an effect on the release mechanism (Baker, 1987). At low drug loading levels (less than 5% by volume), the release of the drug involves dissolution of the drug in the polymer followed by diffusion to the surface of the device. At slightly higher loading levels (5-10% by volume), the release mechanism becomes more complex, since the cavities remaining from the loss of the drug near the surface are filled with fluid from the external medium, and these cavities provide preferred pathways for the escape of material remaining within the device. At these loading levels, the cavities are not connected to form continuous pathways to the surface, but they may increase the overall permeability of the agent in the device (Baker, 1987). When loading of dispersed agent exceeds 20% by volume, the cavities left by the loss of material are sufficiently numerous to form a continuous channel to the surface of the matrix. The majority or the entire active agent is released  58  by diffusion through these channels.  The solubility and diffusivity of the dispersed  agent in the fluid filling the channels determines its rate of release (Baker, 1987). 1.6.3.2 Diffusion and degradation controlled drug release The mechanisms of drug release from biodegradable polymeric devices are more complicated than from non-degradable devices.  The kinetics of drug release are not  only governed by the diffusion of drug from the device but also by the degradation rate of the polymer. The influences of these mechanisms may also vary at different time points of drug release.  Attempts to match release profiles to kinetic equations are usually not  possible because too many variables exist (Chien, 1982). If the drug diffuses from the device rapidly relative to the degradation of the polymer, the drug release from the polymer is mainly controlled by simple diffusion during the initial stages. However, subsequent degradation of the polymer can increase the permeability of the polymer matrix significantly and hence increase drug release rates (Gopferich, 1996a). In degradation-controlled monolithic systems, degradation of the polymer leads to erosion, which is the process of material loss from the polymer bulk (Gopferich, 1996b). Depending on the rate of polymer degradation and water diffusion into the polymer, either surface or bulk degradation and erosion may occur. In the first case, polymer  59  degradation is faster than water diffusion. Thus degradation and erosion are surface phenomena.  In the case of bulk degradation and erosion, water ingress is rapid and  degradation and erosion occurs throughout the polymer marix (Gopferich, 1996b). 1.6.3.3 Swelling controlled drug release A swelling-controlled system consists of a dispersion of a drug in a hydrophilic polymer matrix. In a swelling-controlled system, penetration of water from the environment changes the dimensions as well as the physical properties of the polymer matrix and thus the drug release kinetics. Drug release from such systems is a function of the rate of uptake of water from the surrounding media and the rate of drug diffusion (Baker, 1987). 1.6.4  Factors affecting drug release from biodegradable polymers  A number of factors can affect the kinetics of drug release from a polymer matrix. These include the properties of the drug and the properties of the polymer. In diffusion controlled systems, the molecular weight and solubility of the drug in the polymer and release medium influence the drug release kinetics. The diffusion coefficient of the drug in the polymer decreases as the molecular weight of the drug increases (Pitt and Schindler, 1980).  The solubility of the drug in the polymer matrix  and in the release medium can influence the diffusion of the drug in the polymer matrix  60  and the partition coefficient of the drug between the polymer matrix and release medium. A n increase in the drug solubility in both the polymer matrix and the release medium generally leads to an increase in the drug release rate (Chien, 1982). For degradation controlled systems, acidic or basic drugs can affect polymer degradation through pH changes (Gopferich, 1996a).  For example, amine drugs such as  meperidine and methadone have been shown to increase the rate of polyester hydrolysis (Cha and Pitt, 1989). The degree of crystallinity of a polymer can affect the permeability of the drug molecules in the polymeric matrix. In a semicrystalline polymer, drug diffusion usually occurs in the amorphous regions since the ordered alignment of polymer chains in the crystalline regions lowers the free volume, thus preventing the diffusion of drug and water molecules (Chien, 1982). Therefore, lowering the crystallinity of a polymer will increase its permeability to drugs and water and thus increase the release rate of the drug (Pitt et al, 1979ab).  The flexibility of polymer chains in the amorphous regions also  plays an important role. High flexibility, as indicated by low Tg values, allows the polymer chains to move more readily and thus increase the permeability to drug and water (Pitt et al., 1979ab).  Biodegradable polymers that have low Tg's exhibit faster  biodegradation rates due to the individual polymer molecules possessing more mobility,  61  leading to greater chemical and possible enzymatic attack (Pitt et al., 1979ab).  The  molecular weight of a polymer also affects drug diffusion and polymer degradation.  As  the molecular weight of a polymer decreases, the permeability of the polymer increases due to increased free volume caused by a greater number of polymer chain ends (Cha and Pitt, 1989). For polyesters such as P L A , hydrolysis increases as molecular weight decreases due to a higher percentage of hydrophilic end groups in the lower molecular weight polymer. The increased hydrophilicity permits a more rapid influx of water into the polymer matrix and leads to a higher water content in the matrix (Pitt et ah, 1979ab). Other factors such as drug loading and device geometry can influence drug release. In both monolithic solution and dispersion systems, an increase in drug loading results in an increased concentration gradient, thus increasing release rate (Chien, 1982). In a monolithic dispersion system, the levels of drug loading also influence the formation of cavities and channels for diffusion.  The geometry of the drug delivery system can affect  the diffusion path and surface area of the device. Increasing surface area results in a higher drug release rate (Chien, 1982; Baker, 1987).  62  1.7  INTRAARTICULAR MICROSPHERES AS A DRUG DELIVERY SYSTEM  1.7.1  Microspheres composition and size  Microspheres manufactured using different polymers and in different size ranges have been investigated for their tolerability in joints following intra-articular administration and their suitability for delivering anti-inflammatory drugs to the joints. Using empty microparticles, sized 1-10 pm prepared from poly(lactic acid), poly(butylcyanoacrylate), gelatin, and albumin, Ratcliffe et al (1984) observed various degrees of inflammatory response following intra-articular injection of all polymeric microparticles except albumin.  However, this study failed to investigate the effect of  particle size on the degree of inflammation induced by the different polymers. Greis et al (1994) investigated the interactions of particles with synovial fibroblast cell cultures by measuring the collagenase synthesis induced by particles.  Standard  sized latex beads (0.4, 15, 45, 90 pm) and particles from prosthetic anterior cruciate ligament materials, Dacron and carbon particles, were selected for the study.  The ability  of latex beads to induce collagenase was strongly size dependent. Particles that were 0.4 pm in diameter were readily phagocytosed by the cells and induced collagenase production, while 15 pm latex beads were not readily internalized. Nevertheless, the synthesis of collagenase was induced i f the beads were internalized. 63  Latex particles  with diameters of 45 and 90 u.m were too large for uptake by synovial fibroblast cells and did not induce the synthesis of collagenase.  Both the size and physical properties of  Dacron and carbon influenced their ability to activate synoviocytes. Dacron and carbon particles all induced collagenase secretion.  Internalized  Certain cells that  contained no ingested particles also produced collagenase when in co-culture with cells with ingested particles, indicating that phagocytosis, in addition to inducing collagenase, also induced the release of cell-activating factors which then activated additional cells in the culture (Greis et al., 1994). Nishide et al (1999) investigated the biodegradation and tissue responses to intra-articular poly(D,L-lactic acid) microspheres in the knee joints of rabbits.  Different  sized (< 20 urn - 200 pm) microspheres containing a fluorescent dye were prepared from poly(D,L-lactic acid) polymers with different molecular weights (3000-7000 g/mole). Although there was a temporary increase in the number of white blood cells in the joint, irrespective of the microsphere size, no inflammatory responses to the microspheres by the synovial tissue were observed within 3 days after injection. It was found that following intra-articular injection, the microspheres were localized in the adipose tissues of the popliteal region of the knee cavity, irrespective of the microspheres size and molecular weight (Nishide et al., 1999).  64  Horisawa and coworkers (2002a) investigated the size-dependency of intra-articular nanospheres and microspheres of poly(lactic-co-glycolic acid) (PLGA) on tolerability in rat knee joints.  Histological analysis showed that fluoresceinamine bound P L G A  nanospheres with a mean diameter of 265nm were extensively phagocytosed in the synovium by the macrophages infiltrated through the synovial tissues.  The  phagocytosed nanospheres were delivered to the deep underlying tissues and the synovium was fairly proliferated, 3 days following the injection.  On the other hand,  microspheres with a mean diameter of 26.5 pm were not phagocytosed in the macrophages.  A mild proliferation was observed in the epithelial lining synovial cells  and the microspheres were covered with a granulation of multinucleated giant cells.  The  number of inflammatory leukocytes in the synovial tissue slightly increased one day after the injection of either nanospheres or microspheres, but no further inflammatory responses were detected.  The authors concluded that both nanospheres and  microspheres were well tolerated in the joint (Horisawa et al., 2002a). Using an in vivo isolated horse joint model, Bragdon and coworkers (2001) assessed initial biocompatibility and early (within 3 h) vascular and transsynovial fluid alterations in horse knee joints, following intra-articular injection of paclitaxel loaded P L G A microspheres with an average size of 50 pm. Compared with control (non-injected)  65  joints, intra-articular injection of paclitaxel loaded P L G A microspheres did not affect joint blood flow and blood pressure during this short term study, and early joint reaction was minimal.  Gross and histological morphology of synovium and articular cartilage  were normal in all isolated joints (Bragdon et al., 2001). In our laboratory, experiments have been conducted to investigate the in vivo biocompatibility of microspheres made from different biomaterials and with different particle sizes. Poly(L-lactic acid), poly(lactic-co-glycolic acid) and poly(caprolactone) microspheres with different size ranges (1-20 urn, and 35-105 urn) were injected into joints of healthy rabbits. graded.  The swelling and inflammatory responses were scored and  There was no apparent difference in responses caused by different biomaterials.  The microspheres in the size ranges of 35-105 urn appeared to be well tolerated in the joint while smaller microspheres in the size range of 1-20 urn induced a higher degree of cellular infiltration and proteoglycan loss (Liggins et al., 2004). In summary, previously published works using microspheres made of biodegradable polymers such as P L G A or P L A in size ranges above 20 um have shown that the microspheres were not phagocytosed by macrophages following intra-articular injections and that there were minimal inflammatory responses (Bragdon et al., 2001; Horisawa et al., 2002a; Liggins et al., 2004).  On the other hand, microspheres or nanospheres in  66  sizes less than 10 urn injected intra-articularly, were generally likely to be phagocytosed and in some studies, produced inflammatory reactions such as synovial proliferation, cellular infiltration, and collagenase production (Ratcliffe et al., 1984; Greis et al., 1994; Horisawa et al., 2002a).  1.7.2  Pharmacokinetics and efficacy studies  There are only a few studies that have investigated the pharmacokinetics and in vivo efficacy of drug encapsulated microspheres following intra-articular administration. Ramesh and coworkers (1999) investigated plasma levels of dexamethasone following intra-articular injection of either dexamethasone loaded poly(D,L-lactic acid) microspheres (size 40 - 110 pm) or dexamethasone solution.  The results showed that  approximately 30% of the dose of the dexamethasone solution was detected in the serum 30 min following injection, and thereafter the drug concentration in the serum rapidly decreased over 4 h. No drug was detected in the rabbit serum over 24 h when the rabbit was injected with dexamethasone encapsulated microspheres.  When the microspheres  were collected from the synovial cavity, drug was detected in the microspheres up to 7 days following injection. The authors concluded that most of the drug incorporated in the microspheres was localized in the synovial cavity (Ramesh et al., 1999).  67  Horisawa and coworkers (2002b) investigated the efficacy of intra-articular delivery of betamethasone encapsulated within P L G A nanospheres (300-490 nm) in an antigen-induced arthritis rabbit model.  Monoarthritis was induced in the left knee joint  of the rabbits by the injection of ovalbumin.  Then, betamethasone encapsulated  nanospheres, betamethasone solution or saline were injected immediately after the antigen challenge. The progression of the disease was monitored for 42 days. Compared with betamethasone solution, and saline, the nanospheres formulation showed improved effectiveness in reducing joint swelling and cell infiltration over 21 days.  However, the  levels of proteoglycan and hydroxyproline in the cartilage slightly decreased in the nanospheres treated group compared to the betamethasone solution and saline groups, suggesting possible cartilage damage caused by the nanospheres (Horisawa et al., 2002b). Liggins et al (2004) investigated the efficacy of paclitaxel encapsulated P L A microspheres (35-105 um) in antigen-induced arthritis in rabbits.  Forty milligrams of  control microspheres or 20% paclitaxel loaded microspheres were injected into rabbit knee joints.  Compared with control microspheres, 20% paclitaxel loaded microspheres  produced significantly less joint swelling, cellular infiltration, and cartilage degradation as measured by proteoglycan loss and chondrocyte necrosis over 29 days (Liggins et al.,  68  2004).  It was suggested that paclitaxel loaded P L A microspheres in the size range of  35-105 pm may be a potential formulation for the intra-articular treatment of inflammation in arthritis (Liggins et al., 2004).  1.8  THESIS GOALS AND OBJECTIVES  Intra-articular drug delivery can be very useful for treating rheumatoid arthritis disease flare-ups, synovitis and pain when a small number of joints are affected or for those joints that do not respond to systemic treatments.  Surprisingly, little work has  been done in terms of developing suitable site directed and controlled release intra-articular drug delivery systems to treat inflammatory arthritic conditions. To our knowledge, M T X has not been previously formulated in polymeric microspheres for intra-articular injection. We therefore test the following hypothesis:  M T X loaded polymeric microsphere  formulations based on biodegradable polymers may be developed for intra-articular injection, which are well tolerated, retained in the synovial joint and provide a controlled, localized delivery of M T X into the joint cavity to produce a therapeutic response. goals were to develop and characterize controlled release polymeric microsphere formulations of M T X and to investigate the in vivo biodistribution and efficacy of  69  The  microsphere formulations following intra-articular injection in an established arthritis rabbit model. The development of a controlled release formulation of M T X for use in selected joints would represent a significant advance in rheumatoid arthritis therapy.  A  knowledge of the in vivo efficacy and biodistribution of M T X encapsulated microspheres is fundamental to future work in which this formulation may be optimized for intra-articular controlled release drug delivery in human subjects. The research objectives were: 1) To develop and characterize M T X loaded P L L A microsphere formulations; 2) To determine the in vitro degradation of M T X loaded P L L A microspheres; 3) To evaluate the biocompatibility of M T X loaded P L L A microspheres in healthy rabbit joints; 4) To determine the pharmacokinetics and biodistribution of M T X following intra-articular administration of M T X encapsulated P L L A microspheres in healthy rabbits; 5) To evaluate the in vivo efficacy of M T X loaded P L L A microspheres following intra-articular administration in the antigen-induced arthritis rabbit model.  70  Chapter 2 FORMULATION AND CHARACTERIZATION OF METHOTREXATE ENCAPSULATED POLY (L-LACTIC ACID) MICROSPHERES' 2.1  INTRODUCTION  Poly(L-lactic acid) (PLLA) was selected as a suitable biodegradable polymer for the development of M T X loaded microspheres due to its biodegradability, biocompatibility, mechanical strength and ability to achieve prolonged drug release (Watt et al., 1990). P L L A is commercially available over a wide range of molecular weights from 2k to 100k g/mole.  The ability to prepare microspheres using different molecular weights of  polymers was considered important, so that drug loaded microsphere formulations with significantly different release rate profiles could be developed.  It is well established that  molecular weights of the polymer greatly influences drug release rates (Chien, 1982; Okada, 1997; Siparsky et al, 1998; Tracy, 1998). Other biodegradable and biocompatible polymers, poly(lactic-co-glycolic acid) (PLGA) and poly(D,L-lactic acid) (PDLLA) polymers, along with P L L A , have been used extensively in the formulation of microspheres loaded with a wide range of different drugs (for reviews, see Okada, 1997; Jain et al, 1998; Jain et al, 2000; Tracy et al, 2000  1  A version of this chapter has been published.  Liang LS et al., (2004) Methotrexate loaded poly(L-lactic  acid) microspheres for intra-articular delivery of methotrexate to the joint. J Pharm Sci. 93(4):943-56. 71  and Kakinoki et al., 2003)  Previous work in this laboratory has shown that drug loaded  P L L A microspheres can be prepared with very low molecular weight polymer (around 2k g/mole) and still retain good mechanical integrity (Liggins and Burt, 2001). For this reason, P L L A polymers only were selected for formulation development with M T X . One of the methods commonly employed to manufacture microspheres is the solvent evaporation method.  Figure 2.1 is a schematic diagram showing microsphere  formation by emulsification and solvent evaporation.  The solvent evaporation method  involves the emulsification of a polymer solution containing drug (either dissolved or in suspension) into a second, immiscible liquid phase containing an emulsifier to form a dispersion of drug/polymer/solvent droplets.  The solvent is then removed from the  dispersed droplets by application of heat, vacuum, or by allowing evaporation at room temperature, producing a suspension of drug loaded polymer microspheres that can then be separated by filtration or centrifugation, washed, and dried (Watts et al, 1990). The properties of microspheres can change upon even slight variations of the manufacturing process, resulting in different microstructures (Donbrow, 1992; Gopferich, 1996a). The main variables that influence the microencapsulation process and the final microsphere product, include a) the nature and solubility of the drug being encapsulated; b) the polymer concentration, composition, and molecular weight; c) the drug/polymer ratio; d)  72  the organic solvent used; e) the concentration and nature of the emulsifier used; f) the temperature and stirring/agitation speed of the emulsification process; and g) the viscosity and volume ratios of the dispersed and continuous phase (Jain et al., 1998). The biocompatibility of intra-articular P L L A microspheres has been investigated in healthy rabbit joints (Liggins et al., 2004).  Intra-articular injections of control (no  drug) P L L A (100k g/mole) microspheres in the size range of 35-105 pm, induced mild swelling and cell infiltration (Liggins et al., 2004).  However, P L L A microparticles in  the size range of 1-10 urn induced marked synovial hyperplasia, cellular infiltration and fibrosis in rabbit joints (Ratcliffe et al, 1984). In this study, we evaluated the biocompatibility of control and M T X loaded low molecular weight P L L A microspheres in the size range of 33-110pm.  Rabbits were chosen as a suitable animal model for this  work because the larger joints of rabbits are easier to inject compared to the joints of rats or mice, and rabbits were used in the pharmacokinetic studies in the following chapter. In this chapter, we report the results of development and characterization of poly (L-lactic acid) (PLLA) microspheres loaded with M T X .  The objectives of this study  were: 1. To prepare M T X loaded microspheres using P L L A with three different molecular weights;  73  2.  To characterize the solid state and degradation properties of M T X loaded  microspheres; 3. To determine the in vitro M T X release profiles from microspheres; 4.  To determine the biocompatibility of M T X loaded microspheres following  intra-articular injection in healthy rabbits.  74  2.  Polymer dissolved in organic solvent.  Drug  Emulsifier dissolved in second  added to form a solution  liquid immiscible with  or suspension.  drug/polymer solution.  Polymer/drug solution mixed into 4. emulsifier solution to form a dispersion of drug/polymer/solvent  t  1 1  ©  droplets.  t  © ©  ©  ©  Solvent removed by: Microspheres recovered by filtration a)  Stirring at room temp.  b)  Application of heat.  c)  Application of vacuum.  or centrifugation, then washed and dried.  Figure 2.1 Schematic diagram of microspheres formation by the solvent evaporation method. Adapted from Watts et al (1990).  75  2.2  EXPERIMENTAL  2.2.1  Materials  Methotrexate (MTX) was purchased from Hande Tech Development Co. (U.S.A). Poly (vinyl alcohol) (PVA) (98% hydrated, M W : 13,000-23,000) was obtained from Aldrich Chemical Company Inc. Poly (L-lactic acid) (MW: 2k g/mole, 50k g/mole and 100k g/mole) was obtained from Polysciences (Warrington, PA). Poly (ethylene glycol) standards were purchased from Polymer Laboratories Inc.  Sodium dihydrogen  orthophosphate, sodium phosphate, sodium chloride (NaCl) were purchased from Fisher Scientific (Napean, Ontario, Canada). A l l solvents used were High Performance Liquid Chromatography (HPLC) grade.  Nitrogen and prepurified helium gases were supplied  by Praxair (Burnaby, B C , Canada). Double distilled water was used throughout the studies.  Phosphate buffered saline (PBS, pH 7.4 ) was prepared by dissolving 0.32g  sodium dihydrogen orthophosphate, 2.15 g sodium phosphate, 8.22 g NaCl in one liter of distilled water. The pH of the buffer was in the range of 7.2-7.4.  2.2.2  General equipment and supplies  General equipment and supplies used in this study are listed in Table 2.1.  76  Table 2.1 List of general equipment and supplies used in the study. Equipment and Supplies  Supplier  Balances, Mettler model A E 163, AJ100,  Mettler Instruments, (Zurich, Switzerland)  and PJ 300 Acument Model 230 pH meter  Fisher Scientific, (Fairlawn, NJ)  Olympus BH-2 microscope  Olympus Optical Company Limited, (Japan)  Corning hot plate/stirrer, model PC-351  Corning Glass Works, (Corning, N Y )  GS-6 centrifuge and eppendorf Centrifuge  Beckman, (Palo Alto, C A )  model 5415 C Thelco oven  Precision Scientific Company, (Chicago, IL)  Shel Lab oven  Johns scientific company, (Portland, OR) Precision Scientific Company, (Chicago,  Napco vacuum oven, Model 5831  IL) Caltec Scientific, (Richmond, BC)  Freezer  Scientific Industries, (Bohemia, N Y )  V W R Vanlab vortex mixer  Gilson Company, (Middleton, WI)  Pipettes, variable volume Pipetman Kimax® brand 15mL test tubes with screw caps Buchner funnels  Fisher Scientific, (Toronto, ON)  graduated cylinders Pyrex® brand vacuum dessicator Glass beakers and Erlenmeryer flask Eppendorf tubes  77  2.2.3  , Validation of UV-vis spectrophotometric assay  The UV-vis assays for M T X in PBS were validated by measuring U V absorbance at a wavelength of 304 nm of four sets of standards on three separate days. The linearity, limit of detection, limit of quantitation, inter-day precision, intra-day precision, and accuracy of the calibration curves were measured.  Linearity was expressed as the  coefficient of determination (R ) for each of the twelve calibration curves assembled over 2  three days and for the standard curve consisting of each of the twelve standards' data points for each concentration.  According to the "Reviewer Guidance" published by the  Center for Drug Evaluation and Research, Food and Drug Administration (1994), the limit of detection was calculated from the equation Limit of detection = 3.3a/ s  Equation 2.1  Where a is the standard deviation of the y-intercepts of the twelve calibration curves (four a day for three days) and s is the mean slope of the twelve calibration curves. The limit of quantitation was calculated from the equation Limit of quantitation = 10cr/s  Equation 2.2  Inter-day precision was expressed in terms of the coefficient of variation (CV) of the mean of the twelve standards' data points collected over three days for each concentration.  Intra-day precision was expressed in terms of the C V of the mean of the  78  four standards' data points on each day for each concentration. Accuracy was determined by assembling a calibration curve and comparing three different standards at each concentration against the calibration curve each day for three days. Accuracy was expressed in terms of the percent deviation (bias) of each of the three standards and their mean bias, compared to the daily calibration curve.  2.2.4  Solubility of methotrexate  Three 10 mg samples of methotrexate were placed in 2 mL Eppendorf vials. The powder was dispersed in l m L of pre- warmed double distilled water, 10 m M pH 7.4, phosphate buffered saline (PBS) and 10% poly(vinyl alcohol) solution by vortexing for 10 seconds.  The Eppendorf vials were then placed in a 37°C oven with circular shaking  at a rate of 60 rpm and samples were taken at 24 h. Eppendorf vials were centrifuged at 325 x g for 10 minutes.  The supernatants were diluted with PBS and the absorbances  measured using a UV-vis spectrophotometer (Hewlett Packard Diode Array 8452A) at a wavelength of 304 nm.  2.2.5  Stability of M T X in PBS at 37°C  Three sets of M T X standards in PBS were prepared in the concentration range of 0.4 pg/mL to 50 ug/mL.  The UV-vis absorbances were determined by a U V - V i s  spectrophotometer at a wavelength of 304 nm. The standard M T X solutions were stored 79  in a 37°C oven and the absorbances were measured at time intervals of 5 h, 24 h, and 7 days.  2.2.6  Preparation of M T X loaded microspheres  The microspheres were prepared by the solvent evaporation method.  Five  hundred milligrams of poly(L-lactic acid) with different molecular weights (2k g/mole, 50k g/mole or 100k g/mole ) were dissolved in 2.5 to 5 mL methylene chloride. M T X was suspended in the PLLA7 methylene chloride solution by vortex mixing.  The drug  suspension was then slowly dispersed into 100 mL of a 2.5% P V A solution and stirred at 1000 rpm using an overhead stirrer (BDC2002 Caframo, Ont, Canada). The resultant emulsion was continuously stirred for 2.5 hours at room temperature under ambient pressure until all the methylene chloride had evaporated.  The solidified microspheres  were recovered by centrifugation. The microspheres were sieved through 33 and 110 pm sieves and washed with distilled water.  The washed microspheres were then air-dried  overnight and stored in a desiccator at room temperature for further drying. Either control or M T X loaded microspheres batches intended for injection into rabbit joints were sterilized by gamma irradiation from a Co-60 source at a dose of 25 kGy (Nordion International Inc.).  80  2.2.7  Encapsulation efficiency  To determine the M T X content in microspheres, 5 mg of microspheres were dissolved in 1 mL of methylene chloride in test tubes.  M T X was then extracted by  adding 10 mL of 10 m M , p H 7.4 PBS to the test tube and the tubes were shaken vigorously for 30 seconds.  The drug concentration in the aqueous phase was measured  using a U V - V i s spectrophotometer at a wavelength of 304 nm.  The encapsulation  efficiency was expressed as: (the amount of M T X in the microspheres / the theoretical amount of M T X in the microspheres) x 100%. 2.2.7.1  Validation of encapsulation efficiency studies  To validate the encapsulation efficiency test method, 20 pL, 50 pL or 100 pL of alO mg/mL M T X in a dimethylsulfoxide (DMSO) stock solution were mixed with 100 pL of P L L A (2k g/mole) in a methylene chloride stock solution in a 12 mL glass test tube to give 4%>, 10%o and 20% (w/w) theoretical loadings. The mixture was dried under a stream of nitrogen gas.  The dried mixture was then dissolved with 1 mL of methylene  chloride. M T X was extracted by adding 10 mL of 10 mM, p H 7.4 PBS to the test tube and the tubes were shaken vigorously for 30 seconds.  The drug concentration in the  aqueous phase was measured using a UV-Vis spectrophotometer as previously described.  81  2.2.8  Microspheres characterization  2.2.8.1 Particle size determination The P L L A microspheres (20 mg) were homogeneously dispersed in distilled water (70 mL) with a few drops of 1% polysorbate 80 solution.  Particle size distributions of  microspheres were determined using a Malvern Hydro 2000SM laser diffraction particle size analyzer. Three measurements were taken from each batch and the particle size was expressed as the volume weighted mean. 2.2.8.2 X-ray powder diffraction of microsph eres The X-ray powder diffraction patterns of microspheres and M T X were determined using a Geigerflex X-ray powder diffractometer (Rigaku Inc., Tokyo, Japan). The sample size was approximately 100 mg. The X-ray tube was operated at a potential of 35kV and 18mA.  The range of scans was from 5 to 35 degrees(2f7), and the scan speed  was 1 degree per minute. 2.2.8.3 Scanning electron microscopy Microspheres were mounted on aluminum disks with double sided adhesive tape impregnated with carbon. The mounted samples were coated with 100 A of gold-palladium using a Hummer sputter coater and analyzed by scanning electron microscopy with an electron voltage of 5-10 kV.  82  2.2.8.4 Thermal properties of microspheres Differential scanning calorimetry (DSC) of microspheres was carried out using a Pyris 1 DSC, cooled with liquid nitrogen using a Perkin Elmer Cryofill.  The purge gas  was prepurified helium at a pressure of 20 psi. Microspheres weighing 3-5 mg were placed in crimped, but not hermetically sealed, aluminium pans using an empty pan as a reference.  The samples were heated from -20°C to 200°C at 10°C / min.  The degree of crystallinity (Xc) of P L L A was calculated using the equation: Xc = [(AH - AH )/ 93.7 J/g] x 100% f  C  Equation 2.3  Where AHf and A H are the enthalpies of fusion and recrystallisation, respectively, C  calculated from the area under the curve for both P L L A melting and recrystallisation peaks.  The enthalpy of fusion for 100% crystalline P L L A polymer has been reported as  93.7 J/g (Celli etal, 1992).  2.2.9  In vitro release of M T X from P L L A microspheres  The in vitro release studies were carried out in PBS at 37°C.  Into 12 mL glass,  screw capped tubes were placed 10 mg of M T X loaded microspheres and lOmL PBS. The tubes were tumbled end-over-end at 30 rpm in a thermostatically controlled oven. At given intervals, the tubes were centrifuged at 325 x g for 10 min and 5 mL of the supernatant were saved for analysis.  The remainder of the supernatant was removed and  83  the microsphere pellets were resuspended in fresh PBS (10 mL).  The buffer was  replaced at each sampling interval in order to maintain sink conditions.  The  concentration of M T X in the release medium was measured by U V - V i s spectrophotometry.  2.2.10  Degradation study of microspheres  Into 12 mL glass, screw capped tubes were placed 15 mg of control or drug loaded P L L A microspheres with 10 mL PBS and tumbled at 30 rpm in a 37°C oven. The buffer solution was replaced every day. At predetermined intervals, the microspheres were collected and washed with distilled water three times, dried under vacuum and stored in a desiccator at room temperature. The weights of microspheres samples were then measured to determine the weight loss of microspheres due to degradation. 2.2.10.1  Gel permeation chromatography (GPC)  The molecular weights of degraded P L L A polymer samples were determined using gel permeation chromatography (GPC).  The system consisted of a Waters 515  H P L C pump, a Waters 717 plus autosampler, and a Waters 2410 refractive index detector with a detection cell temperature of 40°C.  The analytical columns were Styrogel® HR3  connected with HR0.5 (Waters Inc.). Assay conditions for the analysis of 0.5 % w/v  84  polymer in tetrahydrofuran were an injection volume of 20 pL, and a mobile phase of tetrahydrofuran flowing at 1 mL/min.  A calibration curve of refractive index versus log  molecular weight was generated by the Millennium software program using poly (ethylene glycol) standards (Polymer Laboratories Inc.)-  The number average  molecular weights and weight average molecular weights of the polymer were calculated using the Millennium software program.  2.2.11  In vivo tolerability of M T X loaded microspheres in rabbit joints  A l l animal experiments were conducted according to the animal care guidelines of the University of Toronto.  The studies were carried out by our collaborators, Drs.  Weixian Ming and Tony Cruz, at Mount Sinai Hospital, Toronto, Ontario. Twelve female New Zealand white rabbits (approximately 2.5 kg) were divided into two groups of 6 animals.  The right knee joint of each rabbit was injected  intra-articularly either with 25mg control P L L A (2k g/mole) microspheres or 10% M T X loaded P L L A (2k g/mole) microspheres in 200 uL PBS.  After the injections, the  animals were observed daily and scored for swelling of joints during the first week. A l l animals were sacrificed on day 14 after injection. The joints were fixed in formalin and then decalcified in 10% formic acid with repeated changes.  The decalcified joints were  paraffin-embedded and joint sections containing synovium, cartilage and bone were  85  prepared.  The sections were stained for cellularity with hematoxylin and eosin or for  proteoglycan content with Saffranin "o". The swelling in joints and the loss of proteoglycans in the cartilage of the joints was evaluated according to the following standard reported previously by Liggins et al (2004): Scoring standard for swelling in joints Grade 0: no swelling Grade 1: little swelling Grade 2: moderate swelling Grade 3: heavy swelling Grade 4:  severe damage and pain (animal cannot walk on the limb)  Scoring standard for loss of proteoglycans in the cartilage by histological analysis Grade 0: no loss Grade 1: small loss (less than 1/3 of total) Grade 2: moderate loss (around 1/3 of total) Grade 3: heavy loss (1/3 to 2/3 of total) Grade 4: total loss  86  2.3 RESULTS 2.3.1  Validation of UV-vis spectrophotometry assay for M T X in PBS  The precision data are given in Table 2.2. Acceptable assay precision was taken to be < 20% for the coefficient of variation value at the lowest concentration measured and < 15% at all other concentrations. These criteria were met using the UV-vis spectrophotometry assay at a concentration range of 0.156 to 40 pg/mL.  A linear  relationship between M T X concentrations in PBS and the U V absorbance at a wavelength of 304 nm in the concentration range of 0.156 ug/mL to 40 ug/mL (r >0.97) 2  was established. The overall inter-day and intra-day coefficient of variation was about 6%. The limit of detection of M T X in PBS was 0.1753 ug/mL, and the limit of quantitation was 0.5313 pg/mL for M T X .  The accuracy data are shown in Table 2.3.  87  Table 2.2 Intra-day and inter-day precision of UV-vis spectrophotometry assay of M T X in phosphate buffer saline (pH 7.4)  MTX concentration  Absorbance  (ixg/mL)  %cv  b  %CV  Absorbance  Absorbance  %CV  at 304 nm  at 304 nm  at 304 nm"  All days  Day 3  Day 2  Day 1  Absorbance  %CV  at 304 nm  0.156  0.008  16.5  0.001  4.5  0.009  7.0  0.010  17.4  0.312  0.014  1.89  0.015  4.4  0.032  7.7  0.024  7.0  0.625  0.033  3.1  0.031  9.6  0.035  10.4  0.034  10.4  1.25  0.065  2.5  0.064  5.8  0.070  10.7  0.067  7.9  2.5  0.134  0.4  0.129  1.8  0.132  3.6  0.133  3.1  5  0.261  1.2  0.257  4.0  0.257  3.6  0.257  4.6  10  0.514  1.6  0.510  5.0  0.510  4.0  0.511  3.9  20  1.020  2.2  1.005  5.7  1.000  7.7  1.008  5.6  40  1.971  3.0  1.962  4.3  1.952  6.8  1.959  5.0  Absorbance at 304 nm is the average calculated from four standard curves at each concentration for each day (n=16 for all days). b  % C V is the coefficient of variation, which is the ratio of the standard deviation to the  average, expressed as a percentage.  A value less than 20% at the lowest concentration  and less than 15% at all other concentrations was taken to indicate sufficient precision.  88  Table 2.3 Accuracy of UV-vis spectrophotometry assay of M T X in phosphate buffer saline solution (pH 7.4)  Day 1  concentration  Measured  (/ug/mL)  value"  a  Day 3  Day 2  MTX  %CV  b  Bias  0  (%)  Measured  %CV  Bias(%)  Measured  %CV  Bias(%)  value  value  0.156  0.11  27.2  -28.8  0.14  14.5  -6.7  0.11  14.1  -27.3  0.312  0.36  4.8  14.8  0.21  9.8  -31.8  0.43  2.4  37.1  0.625  0.42  7.0  -32.2  0.59  6.4  -6.0  0.48  16.4  -22.5  1.25  1.13  3.4  -9.9  1.32  3.2  5.4  1.36  1.5  8.7  2.5  2.58  0.5  3.2  2.72  1.7  8.6  2.73  2.9  9.3  5  5.30  1.8  6.0  5.59  1.0  11.9  5.65  1.2  12.9  10  10.76  1.8  7.5  11.29  1.2  12.9  11.55  1.3  15.5  20  21.74  0.9  8.7  22.45  0.3  12.2  23.43  0.6  17.1  40  42.47  0.2  6.1  43.62  0.9  9.1  45.85  0.6  14.6  Measured value is the average (n=3 for each day) of values of concentration calculated  from one standard curve on that day. % C V is the ratio of the standard deviation to the average, expressed as a percentage.  b  value less than 15% was taken to indicate sufficient accuracy at each concentration. J3ias is the ratio of the deviation of measured value from the actual concentration  c  measured, expressed as a percentage.  89  A  2.3.2  Solubility and chemical stability of M T X  The solubilities of M T X at 37°C in double distilled water, PBS (pH 7.4) and 2.5% P V A were 0.15, 6.71 and 0.47 mg/mL, respectively. The chemical stability of M T X in PBS at 37°C over a concentration range from 0.156 to 40 pg/mL over a week is summarized in Table 2.4. The percentage degradation of M T X was determined by comparing the absorbance at 304 nm of the same sample measured by the UV-vis spectrometer on the day of sample preparation and after 7 days at 37°C.  The results showed that at low M T X concentrations (0.156- 0.625 u.g/mL) the  percentage M T X degraded was more than 20%, while the degradation for higher concentrations (1.25 - 40pg/mL) was less than 10%. Table 2.4  Percentage degradation of M T X in phosphate buffered saline (pH. 7.4)  following 7 days of storage at 37°C.  The values are mean ± standard deviation (n=4).  M T X concentration  Mean degradation (%)  (Ug/mL) 0.156  27.0±7.1  0.312  28.6 ± 2 . 1  0.625  25.5 ± 0 . 5  1.25  9.4 ± 0.4  2.5  9.1 ± 0 . 6  5  11.4 ± 0 . 6  10  5.4 ± 1.3  20  9.2 ± 0.4  40  5.7 ± 0 . 2  90  2.3.3  Optimization of manufacturing of M T X loaded microspheres  Poly (L-lactic acid) of three different molecular weights (2k, 50k and 100k g/mole) (PLLA2k, PLLA50k, and P L L A 100k) were used to manufacture M T X loaded microspheres.  When stirring rate and the concentration of emulsifying agent remained  constant, the effects of changing polymer concentration and M T X loading on the encapsulation efficiency and particle size of the microspheres were evaluated and are shown in Table 2.5. For PLLA2k microspheres, when the polymer concentration was increased from 10% to 20%, the encapsulation efficiency increased from 35% to 68% and the mean particle diameter increased from 28 um to 83 um.  Therefore, for further  characterization, all PLLA2k microspheres were manufactured using 20% polymer concentrations.  The microspheres exhibited unimodal particle size distributions as  illustrated in Figure 2.2. The microspheres prepared from P L L A 2k had smaller mean particle sizes (62-83 urn) while the microspheres prepared from high molecular weights (PLLA 50k and 100k) showed larger mean particles sizes (greater than 100 urn). The manufacturing process allowed for high levels of drug encapsulation with all three polymers. PLLA2k microspheres produced between 64-89%) M T X encapsulation efficiency while PLLA50k and P L L A 100k microspheres showed more than 80% M T X encapsulation efficiency. To ensure that the encapsulation efficiency test could  91  correctly determine the amount of M T X encapsulated in the microspheres, PLLA2k and M T X solutions of known concentrations were dispensed into a test tube and dried out using nitrogen gas.  The dried P L L A - M T X suspension was processed the same way as  for the encapsulation efficiency test and the amount of M T X in the suspension was determined.  The results showed that when the theoretical loading was 5% and 10%, the  recovery of M T X was approximately 96%. The recovery of M T X dropped to 88% when the theoretical loading was increased to 20%. A l l microspheres for further characterization studies were in the size range of 33 pm- 110 pm. The size range was controlled by sieving microspheres through sieves with mesh opening of 33 pm and 110 pm. The percentage yields of microspheres following sieving are given in Table 2.6. For P L L A 2k microspheres, approximately 60% of microspheres manufactured were in the size range of 33-110 pm while for PLLA50k and 100k, 70-80%) of total microspheres manufactured were in this size range.  92  Table 2.5 Manufacturing conditions and properties of M T X loaded microspheres manufactured from 2k, 50k and 100k g/mole P L L A polymer. Polymer  Theoretical Encapsulation Mean  Polymer  efficiency (%) diameter at 90th  molecular concentration loading weight  (w/v %)  Diameter  (w/w %)  (pmf  (g/mole)  percentile  (M*n) 83.7+0.5 128.5+0.5  20  0  n/a  20  2.5  63.6  79.5±0.8  128.1+1.2  20  5  65.6  61.6±0.4  101.3+0.1  20  10  68.0  76.910.7  130.5+1.9  10  10  35.3  28.7+0.4  70.5+0.4  20  20  89.5  77.5±0.5  127.8+0.4  50k  10  10  89.6  104.3+0.4  179.0+0.5  100k  10  10  82.5  187.6+0.6  325.3+1.7  2k  a  The volume weighted mean is reported as the mean diameter values. The values are  shown as the mean of 3 batches of microspheres ± one standard deviation (n=3).  93  Table 2.6  Percentage yield of microspheres manufactured from PLLA2k, 50k and 100k  following sieving between 33 um, and 110 um sieves. PLLA  molecular  weight  Theoretical loading of MTX  % yield of 33-110 nm  (w/w %)  microspheres'  10  57.33 ±2.35  Control  59.71 ±5.95  10  67.64 ± 1.93  Control  68.68 ±2.88  10  80.09 ± 0.66  control  77.27 ± 2 . 9 7  1  (g/mole) 2k 50k 100k a  % yield of 33 -110 um microspheres was determined as the percentage weight of  microspheres in the size range of 33 -110 um divided by the theoretical weight of microspheres which is the weight of M T X plus the weight of polymer.  94  Particle Size Distribution  12 10 8 >lum  CD  >  6 4  2 01  0.1  10  1000  3000  Diameter (um)  Figure 2.2  Particle size distributions of 10% M T X loaded P L L A microspheres  manufactured from 2k, 50k, and 100k g/mole P L L A polymer using the solvent evaporation method.  95  2.3.4  In vitro drug release profiles  2.3.4.1  The effect of polymer molecular weight on MTX release  Figure 2.3 shows the in vitro release profiles of M T X loaded microspheres in the size range of 33-110 pm as a function of molecular weight. Microspheres prepared from 2k, 50k, and 100k P L L A all demonstrated a rapid burst phase of release of M T X followed by a slower release period. More than 50% of the drug was released from P L L A 2k microspheres within the first day followed by a slow release period in which about 10% of the drug was released in the remaining 14 days.  The burst phase of release  for PLLA50k and PLLAlOOk microspheres produced less than 20%) of the total drug being released over the first day. Following the burst phase, only about 4 % of the drug was released over the remaining 35 days. Approximately 95% of the total M T X was released over 20 days from PLLA2k microspheres, while 27% and 24% of total M T X was released from PLLA50k and PLLAlOOk microspheres, respectively, over 40 days. M T X loaded microspheres prepared using PLLA2k were selected as the lead formulation for future characterization work, based on the faster and more complete release of M T X from the P L L A 2k microspheres over a period of 3-4 weeks, compared to the PLLA50k and PLLAlOOk microspheres.  96  2.3.4.2 The effect of MTX loading on MTX release The effects of drug loading on M T X release was further studied in PLLA2k microspheres.  Figure 2.4 shows the in vitro release profiles for M T X from  microspheres prepared from PLLA2k loaded with 4 different amounts of M T X . PLLA2k microspheres loaded with 2.5, 5, 10 and 20 % (w/w) M T X all showed a burst phase of release followed by a much slower phase of release over 14 days. detailed release profile within 24 h is also shown (Figure 2.4B).  The  The amount of drug  released was dependent on the amount of M T X loaded in the microspheres. Approximately 60% of the loaded amount of M T X was released in the burst phase followed by a phase giving approximately zero order release at rates of 1.34, 2.36, 2, and 4.5 pg/day for 2.5, 5, 10 and 20% (w/w) loaded microspheres, respectively, over the remaining 20 days. 2.3.5.3 The effect of gamma irradiation on drug release Cumulative M T X release profiles for non-irradiated and y-irradiated PLLA2k microspheres loaded with M T X are shown in Figure 2.5. It showed that irradiated microspheres have release profiles almost identical to non-irradiated microspheres.  97  800  r  700 Y  0  10  20  30  40  50  Time (d)  Figure 2.3  Cumulative in vitro release profiles of M T X in 0.1M PBS at 37°C from  microspheres loaded with 10% (w/w) M T X manufactured from P L L A polymers with molecular weights of 2k, 50k and 100k g/mole.  Microspheres were in the size range of  33- 110 pm. Values are mean ± one standard deviation, (n=4). (x) PLLA2k, ( A ) PLLA50k, (•) PLLAlOOk.  98  A) 1600  r  Time (h)  Figure 2.4  Cumulative in vitro release profiles of M T X in 0.1 M PBS at 37°C from A)  microspheres loaded with various amounts of M T X manufactured from P L L A 2k g/mole Microspheres were in the size range of 33-110 um. The detailed release profile within 24 h is shown in B).  Values are mean ± one standard deviation, (n=4). M T X loadings  were (x) 20%(w/w), ( • ) 10% (w/w), (•) 5% (w/w), (•)2.5% (w/w).  99  100  r  0  1  2  3  4  5  6  7  8  9 10 11 12 13 14  Time (d)  Figure 2.5  Cumulative in vitro release profiles of M T X in 0.1 M PBS at 37°C from  y-irradiated (25 kGy from Co-60 source) and non-irradiated PLLA2k microspheres loaded with 10% M T X (w/w). Microspheres were in the size range of 33- 110 pm. Values are mean ± one standard deviation, (n=4). microspheres samples.  100  (•) non-irradiated, (•) irradiated  2.3.5  Characterization of microspheres  2.3.4.2 Surface morphology  The scanning electron microscopic surface morphologies of both control (no drug) and 10% M T X loaded PLLA2k microspheres are shown in Figure 2.6.  Both  control (A) and 10% loaded microspheres (B) appeared to be smooth and spherical. There were no M T X crystals precipitated on the surface of M T X loaded microspheres. 2.3.4.3 X-ray powder diffraction patterns of MTX and PLLA microspheres  Figure 2.7 shows the X-ray powder diffraction patterns of crystalline M T X (A) and PLLA2k microspheres (B and C).  Both control (B) and 10% M T X loaded PLLA2k  microspheres (C) had similar X-ray powder diffraction patterns, with the most intense peak located between 16.4 and 16.8°20 and two less intense diffraction peaks at 15 and 19 °2d.  No peaks could be attributed to crystalline M T X in the microsphere matrix from  the X-ray powder diffraction pattern of 10% M T X loaded microspheres. 2.3.4.4  Th ermal properties  Figure 2.8 (A and B) shows DSC thermograms of control and 10% (w/w) M T X loaded PLLA2k microspheres and Table 2.6 provides a summary of the thermal data. The solidification of PLLA2k in the microspheres resulted in a semicrystalline polymer matrix with a degree of crystallinity of approximately 5% (Table 2.7).  The amorphous  component of the P L L A matrix showed a Tg of 52°C, which was taken as the peak 101  temperature of enthalpy relaxation (Figure 2.8A).  Above the Tg, an exotherm was  observed at 95°C and was taken as the crystallization temperature (Tc). Further heating of the sample resulted in melting of the crystalline regions of the polymer matrices, with some evidence of double melting endothermic peaks (Figure 2.8A).  The peak of the  second melting peak was taken as the melting point, which was about 145°C for both control and 10% M T X loaded P L L A 2k microspheres.  The Tg and Tm of the PLLA2k  microspheres were not significantly affected by the addition of 10% M T X (Figure 2.8B).  102  B)  A)  Figure 2.6  The surface morphology of A) control and B) 10% (w/w) M T X loaded  PLLA2k microspheres.  103  Figure 2.7 X-ray powder diffraction patterns of A) crystalline M T X (from Handetec), B) control PLLA2k microspheres in the size range of 33-110 pm, C) 10% M T X loaded PLLA2k microspheres in the size range of 33-110 pm.  104  0  60  r  1  -50  1  0  1  1  50  100  1  1  150  1  200  250  Temperature (°C)  Figure 2.8  DSC thermograms of control and 10% M T X loaded P L L A (2 kg/mole)  microspheres following degradation in 0.1 M PBS at 37°C. A ) Control microspheres, B) 10% (w/w) M T X loaded microspheres before degradation, C) 10% M T X loaded microspheres 1 day, D) 10 % M T X loaded microspheres 28 days following degradation. Samples (3-5mg) were heated from -20°C to 200°C at 10°C /min. (Endotherms down)  105  Table 2.7  Thermal properties of control and 10% M T X loaded P L L A 2 k microspheres  following in vitro degradation in 0.1m PBS (pH 7.4) at 37°C.  Microspheres are in the  size range of 33-110 urn.  10% M T X loaded microspheres  Control microspheres Degradation  Tg  Tc  time (days)  (°C)  0  52  97  1  54  3  Tm  Xc  Tg (°C)  146  (%)* 5±3  100*  147  54±1  101*  7  55±1  14  Tm  Tc  Xc  (°C/' '  (°C)  51  95  145  (%)* 4  9+2  53±2  100*  147  11±6  147  7±2  53  101*  147  13+3  101*  147  14±2  55±2  100*  147  13  53+1  101*  147  11±4  53+1  101*  147  13+3  21  51  100*  145  10±5  52  100*  146  9±5  28  57±1  100*  145  12+2  55±2  100*  146  10±4  a  (°C) ' fl  c  (°C)  a  fl  C  fl  Tg: glass transition temperature, Tc: crystallization temperature, Tm: melting temperature, X c : degree of crystallinity "The standard deviation of the mean temperatures is not shown i f less than 1°C. *The standard deviation of the mean degree of crystallinity is not shown i f less than 1%. c/c  Significantly different by one-way A N O V A test (FO.0001) among Tc values for  microspheres after days of degradation. * Significantly different from day 0 by Tukey-Kramer test. Values are mean values ± standard deviation, (n=3).  106  2.3.6  Degradation of microspheres  A n in vitro degradation study in PBS at 37°C was conducted for PLLA2k microspheres using GPC determination of molecular weight changes over time. A calibration curve of log molecular weight versus retention time using poly(ethylene glycol) standards is shown in Figure 2.9.  Figure 2.10 shows the number average  molecular weight and weight average molecular weight changes for the PLLA2k microspheres incubated in PBS at 37°C.  There was no difference between the molecular  weights of drug loaded and control microspheres over the degradation time period. The results showed that there was an apparent increase in the molecular weights of the polymer for both number average molecular weight and weight average molecular weight after one day of degradation.  Despite the initial increase in molecular weight, by  30 days of degradation, P L L A 2k microspheres had lost approximately 20% of the original molecular weight (Figure 2.10). The mass loss profile (Figure 2.11) shows that approximately 13% of total mass of 10% drug loaded microspheres was lost after one day of degradation while 9 % of total mass of control microspheres was lost after one day of degradation.  The mass of the  microspheres decreased in a linear fashion with the time of degradation.  After 35 days  of incubation, approximately 22 % of the total mass was lost due to degradation.  107  5 -i  1.5 1 -| 10  1  12  1  1  14  16  1  18  1  20  Retention time (min)  Figure 2.9  Molecular weight calibration curve using poly(ethylene glycol) standards.  Chromatographic conditions: Styrogel® HR3 and HR0.5 columns in series, mobile phase was THF at 1 mL/min, and the detector was a differential refractive index detector.  108  A)  43  '53  M "3 o  u  o S  IP o> > (3  <u  4500 4000 3500 3000 , „ 2500 « ° 2000 ^ 1500 1000 500 H 0  15  10  20  25  30  Time (d)  B)  Time (d)  Figure 2.10  A ) Number average molecular weight and B) weight average molecular  weight profiles of control and 10% (w/w) M T X loaded PLLA2k microspheres in 0.1 M PBS at 37°C.  Microspheres were in the size range of 33- 110 pm. Values are mean ±  one standard deviation, (n=3). (•) 10% M T X loaded, (•) control microspheres.  109  Figure 2.11 Cumulative weight loss profile of control and 10% M T X loaded PLLA2k microspheres incubated in 0.1M PBS at 37°C.  Microspheres were in the size range of  33- 110 um. Values are mean ± one standard deviation, (n=3). (•) 10% M T X loaded, (•) control microspheres.  110  Figure 2.12  Scanning electron micrographs of control and 10% (w/w) M T X loaded  PLLA2k microspheres following degradation in 0.1 M PBS at 37°C. Microspheres were in the size range of 33- 110 pm.  A) Day 3, control, B) day 3, 10% M T X , C) day 7,  control, D) day 7, 10% M T X .  Ill  2.3.6.1 Thermal analysis of degraded microspheres Representative DSC thermograms of degraded microspheres (day 3 and day 25) are shown in Figure 2.8 (C and D).  The thermograms showed the disappearance of the  enthalpy relaxation endotherm and an elevation of Tc by 5°C (Figure 2.8 C and D). The Tg values of the degraded microspheres were taken as the mid-point in the change in heat flow.  The changes in Tg, Tc, and Tm of the microspheres following degradation are  summarized in Table 2.7. Both control and 10% M T X loaded microspheres showed a significant elevation in the values of Tc following degradation while the Tg and Tm were not affected by degradation.  The crystallinity was observed to increase from 5% to 11%  after one day of degradation although the change was not statistically significant. 2.3.6.2 Surface morphology of degraded microspheres The surface morphology of degraded microspheres is shown in Figure 2.12. On day 3, both control and drug loaded microspheres exhibited some evidence of degradation (Figure 2.12, A and B) with the appearance of some pores and holes in the surface.  On day 7, the M T X loaded microspheres showed greater degradation and loss  of integrity (Fig 2.12, D) compared to control microspheres (Fig 2.12, C).  112  2.3.7  Biocompatibility of M T X loaded and control microspheres  Healthy rabbits were injected intra-articularly with 25 mg of either control or 10% M T X loaded P L L A 2 k microspheres.  After the injection, animals were observed daily  and scored for swelling of joints during the first week.  The results are given in Table  2.8. In the first 3 days, little to moderate swelling was observed in both groups.  The  swelling reaction decreased three days after the injection. There was no swelling observed in the remaining 4 days. Fourteen days after the injection, the animals were sacrificed and the joints were taken for histological study.  Both safranin-0 stain and toluidine blue stain showed that  the animals from both groups had a small degree to no loss of proteoglycans from the cartilage (Table 2.8).  The results indicated that the 10% M T X loaded and control  PLLA2k microspheres in the size range of 33-110 um were well tolerated in the joints of the rabbits.  113  Table 2.8  Joint swelling and histological analysis of proteoglycan loss in cartilage of  rabbit joints injected intra-articularly with 25 mg control or M T X loaded PLLA2k microspheres in the size range of 33 -110 pm. Joint swelling and proteoglycan loss were scored on a 0 to 4 point scoring system. Treatment /Animal  Loss ofproteoglycan  Scores for swelling of joint  Number  Control  Dayl  Day2  Day3  Day4  Day5  Day6  Day7  Safranin"0"  Toluidine  stain  blue stain  Microspheres 1  0.5  1  0  0  0  0  0  1  0.5  2  0.5  1  0  0  0  0  0  0  0  3  1  1  1  0  0  0  0  1  1  4  1  2  1  0  0  0  0  1  1  5  1  1  1  0  0  0  0  0.5  0  6  0.5  1  1  0  0  0  0  0  0  1  1  1  0  0  0  0  0  0.5  0.5  2  1  1  0  0  0  0  0  1  1  3  0.5  0  0  0  0  0  0  0  0  4  1  1  0  0  0  0  0  1  1  5  2  1  0  0  0  0  0  0.5  0  6  2  1  0  0  0  0  0  1  1  MTX Microspheres  "Control microspheres" animals were treated with 25 mg PLLA2k microspheres and " M T X microspheres" animals were treated with 25 mg 10% M T X loaded PLLA2k microspheres.  114  2.4 D I S C U S S I O N  The solvent evaporation method was used to encapsulate M T X in P L L A microspheres.  The data showed that M T X may be encapsulated at relatively high  efficiency (between about 64-90%) in microspheres manufactured from P L L A of various molecular weights (Table 2.5).  The lower M T X encapsulation efficiency achieved for  the low molecular weight polymer (2k g/mole) compared to the high molecular weight polymer (50k and 100k g/mole) was likely due to the difference in the precipitation rates of the different polymers during the microsphere formation process.  To determine  relative rate of polymer precipitation and hardening during microspheres formulation, samples of polymer droplets were taken every 5 min and observed using optical microscopy.  It was noted that high molecular weight P L L A precipitated and hardened  as microspheres more rapidly than PLLA2k.  Within 5 to 10 min, PLLA50k, and 100k  polymers precipitated and hardened as microspheres, while PLLA2k still appeared as emulsion droplets in the water phase up to 30 min.  Given that the solubility of M T X in  2.5% P V A solution was found to be approximately 0.5mg/mL (Section 2.2), the slower precipitation rate of PLLA2k and longer hardening time would provide a longer time for M T X suspended in the polymer solution droplets to partition into the aqueous phase resulting in a lower encapsulation efficiency.  115  Particle size data shown in Table 2.5 showed that particle size distributions were reproducible between batches made with a given stirring rate and P V A concentration. The increase in mean particle sizes of the microspheres observed with increasing polymer molecular weight (Table 2.5 and Figure 2.2) may be explained by the difference in the viscosities of solutions of polymers with different molecular weights in methylene chloride. It has been reported that the intrinsic viscosities of 100k g/mole P L L A , 50k g/mole P L L A and 2k g/mole P L L A are 1.067 +0.016, 0.855 + 0.003, and 0.153 ± 0.011 dL/g, respectively (Liggins, 1998). For a given rate of stirring, higher viscosities of polymer in the organic phase generally result in an increased resistance of the organic phase droplets to shear stress and break up, resulting in larger microspheres (Jail and Nixon, 1990; Freiberg and Zhu, 2004).  Particle size of microspheres is known to be a  factor influencing the drug release rates and solid state properties of microspheres (Jali and Nixon, 1989; Jalil and Nixon, 1990) and, therefore, only microspheres in the size range (33-110 pm) intended for intra-articular injection were selected for release study and further characterization. The in vitro M T X release studies of M T X loaded microspheres composed of P L L A with three different molecular weights were conducted in order to select a potential lead formulation for further characterization work. In vitro M T X release profiles from M T X  116  loaded microspheres manufactured from PLLA2k, PLLA50k, and PLLAlOOk showed that the rate and extent of release were dependent on polymer molecular weight (Figure 2.3).  The release rate and the extent of release was highest for PLLA2k microspheres  followed by PLLA50k and PLLA100k microspheres.  There was an initial burst phase  of M T X release followed by a slower and controlled release of the drug for all microspheres.  The high molecular weight polymer microspheres released M T X very  slowly after the burst phase and only 20% of the loaded drug had been released in 40 days.  There was no evidence of matrix erosion on day 7 following drug release from  PLLA50k and PLLAlOOk microspheres indicating that the major factor contributing to M T X release from high molecular weight P L L A microspheres was diffusion rather than erosion (data not shown).  It has been reported that P L L A with molecular weights of  103k g/mole and 153k g/mole showed negligible weight loss during the first 270 days of incubation in Ringer's solution at 37°C (Migliaresi et al, 1994).  The high molecular  weight P L L A microsphere formulations were not felt to be optimal formulation for future in vivo work because the very low release rates of M T X could result in sub-therapeutic tissue concentrations of drug particularly in the acute antigen-induced arthritis model. The long degradation lifetime of high molecular weight P L L A microspheres could be of potential concern for long-term tolerability and biocompatibility in the joint.  117  Hence,  M T X loaded P L L A 2 k microspheres were selected as the lead formulation for subsequent studies. The thermal analysis of PLLA2k microspheres showed that the solidification of P L L A (2k g/mole) resulted in a semicrystalline polymer matrix with a degree of crystalllinity around 5% for both 10 % M T X loaded and control microspheres (Table 2.7).  Glass transitions (Tg) observed in control or 10% M T X loaded microspheres  exhibited concurrent enthalpy relaxation (Figure 2.8 A and B). occur upon heating a polymer through its Tg.  Enthalpy relaxation may  Below Tg the polymer chains may relax  by undergoing short range motion and becoming ordered over the range of a few monomer units of polymer chains. The region of order is not large enough to be considered crystalline and has been termed "microstructure" in the amorphous phase (Bodmeier et al., 1989). Upon heating through Tg, an enthalpy of relaxation is observed because energy is required to overcome the short-range order established through relaxation (Bodmeier et al., 1989; Liggins and Burt, 2001). The peak attributed to enthalpy relaxation at the Tg has been reported for other P L L A polymers (Migliaresi et al., 1994; Gonzalez et al, 1999; Liggins and Burt, 2001). Above Tg, a crystallisation exotherm was observed as the polymer chains gained more mobility and were able to align into a more stable configuration by releasing energy.  118  The peak of the  crystallization exotherm was taken to be the Tc. Further heating resulted in melting of the crystalline regions of the polymer matrices with an evidence of double melting endotherms (Figure 2.8 A and B).  Based on the melting and recrystallization model of  multiple melting of polymers, as the initial crystalline lamellae in the polymer matrix melt and give rise to the low temperature endotherm, the molten material can undergo a recrystallization process during the DSC scan and form thicker lamellae (Wang et al., 1999).  The recrystallized lamellae melt at a higher temperature and result in the  endotherm at a higher temperature (Wang et al, 1999).  The double melting  phenomenon has been observed in some P L L A polymers before and following degradation (Gonzalez et al, 1999; L i , 1999; Martin et al, 1999). The Tg and Tm of 10% loaded P L L A 2k microspheres were not significantly affected by the addition of M T X (Table 2.7) likely due to the fact that drug had negligible solubility in the P L L A matrix.  In general, when a drug is soluble or miscible in a  polymer matrix, the polymer chains are either stiffened by interactions with the drug molecules resulting in a elevation of Tg or they are plasticized by interruption of polymer-polymer interactions by the presence of drug molecules resulting in a depression of Tg (Mumper et al, 1992; Liggins 1998).  Thus, M T X was likely dispersed as solid  particles in the polymer matrix. M T X has a low solubility in methylene chloride of  119  approximately 30 pg/ml (preliminary study).  During microspheres preparation, the  organic phase was a suspension of M T X in a PLLA/methylene chloride solution. This observation and DSC evidence suggests that the dispersion of M T X in the P L L A microspheres matrix was likely a particulate dispersion. The X-ray powder diffraction pattern of M T X , as received, indicates that the drug was crystalline in nature (Figure 2.7). However, no peaks could be attributed to crystalline M T X in the microsphere matrix from the X-ray powder diffraction pattern of 10% M T X loaded microspheres (Figure 2.7) likely due to the low sensitivity of the X-ray powder diffraction technique and inability to detect crystalline M T X at low loadings. The solid state characterization of crystalline M T X conducted by Chan and Gonda (1991) has shown that crystalline M T X was hydrated with 4% of water.  The dehydration of  M T X took place over a temperature range of 40°C to 120°C and the melting of M T X occurred in a temperature range of 175°C to 200°C accompanied by decomposition of M T X (Chan and Gonda, 1991).  We confirmed these observation by DSC analysis of  M T X , as received (data not shown).  Thermal analysis of 10% M T X loaded PLLA2k  microspheres did not detect any peak for M T X melting. We speculate that particulate M T X in the polymeric matrix could have gradually dissolved in the polymeric matrix  during DSC analysis at temperatures above Tm (145°C) of the P L L A resulting in disappearance of the melting peak of M T X . The effects of M T X loading on drug release were investigated in PLLA2k microspheres.  The in vitro drug release profiles of PLLA2k microspheres loaded with  2.5%, 5%, 10%o and 20% M T X all demonstrated a burst phase of release followed by a much slower phase of release over 14 days (Figure 2.4). M T X loading levels affected the rate and extent of M T X release. loading levels of M T X .  The rate of M T X release was increased with higher  According to the kinetics described for diffusion controlled  release from a sphere by Baker (1987), the amount of drug released at any given time increases directly proportionally to the total drug loading.  The initial burst phase of  M T X release was likely due to dissolution of M T X near the surface of the microspheres. Since M T X in P L L A microspheres matrix was likely dispersed as solid particles, as the release medium penetrated the microsphere matrix, it would dissolve particulate M T X , leaving cavities and pores that were filled with release medium and served as a preferred pathway for drug diffusion, contributing to a rapid burst release phase.  The initial stages  of formation of cavities and pores were demonstrated by the scanning electron micrographs of M T X loaded microspheres following 3 days of drug release (Figure 2.12B).  The slower release phase following the burst phase was likely controlled by a  121  combination of drug diffusion and polymer degradation.  The scanning electron  micrographs of M T X loaded and control PLLA2k microspheres by day seven of incubation in PBS showed substantial evidence of polymer erosion (Figure 2.12 D). The effect of gamma irradiation on drug release from M T X loaded PLLA2k microspheres was investigated.  The results showed that the release profiles of M T X  loaded microspheres for 14 days were not altered upon exposure to gamma irradiation at a dose of 25 Gy (Figure 2.5).  Gamma irradiation has generally been employed for  sterilization of biodegradable polymer devices although the dramatic decrease in polymer molecular weight on treatment is a well-known concern (Scholes et al., 1997).  The  effects of gamma irradiation on the properties of polymeric microspheres have been widely investigated (Montanari et al, 1998; Blanco et al, 1999; Montanari et al, 2001; Wang et al, 2003; Faisant et al, 2003). In general, gamma irradiation has a dose dependent effect on the molecular weight and thermal properties of the polymer (Calis et al, 2002; Martinez-Sancho et al, 2004). It has been found that the drug release rate from microspheres increased as the dosage of gamma irradiation increased, but seemed to be less sensitive to the effect of gamma irradiation compared to polymer molecular weight (Calis et al, 2002; Faisant et al, 2003). Other studies have reported that the drug release profiles were not significantly influenced by gamma irradiation even though  122  the molecular weight of the polymer was decreased by chain scission of the polymer chains by gamma irradiation (Martinez-Sancho et al., 2004). The molecular weights of control and 10% M T X loaded microspheres following degradation were monitored by GPC.  The number average molecular weight of  PLLA2k polymer before degradation was determined to be 3500 g/mole (Fig 2.10A), which is higher than that reported by the manufacturer.  This difference is likely due to  the different methods used to measure the molecular weight of the polymer. The manufacturer used an end group titration method, while GPC was used to evaluate the molecular weight in this work. Since no P L L A standards were available, poly(ethylene glycol) (PEG) standards were used for GPC calibration. GPC separates according to the hydrodynamic volumes of the polymer coils in solution and not according to the absolute molecular weights, and hence the GPC values should only be seen as providing relative molecular weights for comparative purposes (Hakkarainern et al, 1996). One day following the degradation, there was an increase in the molecular weight of the polymer (Fig 2.1 OA and B), which was likely due to the rapid leaching of a very low molecular weight fraction of the polymer matrix, resulting in an increase in the average molecular weight of the polymer. The mass loss study also showed that approximately 9% and 14% of total mass was lost following one day of degradation for control and 10%  123  M T X loaded microspheres, respectively (Figure 2.11).  Liggins and Burt (2001)  reported that approximately one-third of l k g/mole P L L A was water soluble and GPC analysis of the water soluble fraction of low molecular weight P D L L A showed that the highest molecular weight that dissolved in water was approximately 650 g/mol (Gradfils et al., 1996). It has also been shown that the average molecular weight of poly (L-lactic acid) films increased from 22k g/mole to 30k g/mole during the first week of degradation in a mineral medium at 37°C due to the extraction of low molecular weight products that became soluble in the mineral medium (Hakkarainen et al., 2000).  Starting on day 3,  the molecular weight of PLLA2k microspheres decreased continuously (Figure 2.10 A and B) and approximately 20% of the total weight of microspheres was lost following 40 days of degradation (Figure 2.11) suggesting that the degradation of microspheres was accompanied by erosion. Thermal analysis of degraded PLLA2k microspheres showed that Tc was elevated and crystallinity of the microspheres was increased one day following degradation (Table 2.7).  The degradation process starts with a random chain scission process as the ester  bonds are hydrolyzed by the water molecules.  When the molecular weight drops  sufficiently by hydrolysis, the degradation products become soluble or separate from the matrix, and loss of material will occur (Jalil and Nixon 1990).  124  Hydrolysis usually takes  place in the amorphous regions of the polymer matrix due a higher permeability of the aqueous medium to the matrix and the susceptible functional groups in the polymer backbone.  Therefore, degree of crystallinity of polymers generally is increased during  the initial hydrolysis of the polymer due to the preferential hydrolysis of the amorphous regions (Hakkarainen et ah, 2000). It has been shown that following 110 weeks of degradation in phosphate buffer (pH 7.4) at 37°C, the crystallinity of high molecular weight P L L A (130k g/mole) increased by 50% and the Tc of the polymer was elevated by 2°C (Li, 1999; Vert et al., 1994). Thermal analysis of degraded microspheres did not show significant changes in Tg and Tm values of the microspheres even though the molecular weight was decreased 25% after 28 days of degradation (Table 2.7 and Figure 2.11). Tm and Tg values have been reported to be less sensitive to chain cleavage and molecular weight reduction than the degree of crystallinity (Migliaresi et al., 1994). Biocompatibility or tolerability studies in healthy rabbit joints were carried out using control PLLA2k and 10% M T X loaded PLLA2k microspheres in the size range of 33-110 um. This size range was chosen based on previous studies which showed that P L L A microspheres in this size range were well tolerated in rabbit joints, while microspheres in the smaller size range frequently induced a greater inflammatory response (Ratcliffe et al, 1984; Liggins et al, 2004). Intra-articular injections of 25 mg  125  of microspheres (control or M T X loaded) produced a mild inflammatory response over the first three days and subsided thereafter without any significant loss of proteoglycans from cartilage 14 days following injection (Table 2.8). These findings were consistent with our previous work (Liggins et al, 2004) which showed that intra-articular microspheres in the size range of 33-105 um only induced mild knee joint swelling and cellular infiltration seven days following administration of a 40 mg dose.  Nishide and  coworkers (1999) investigated the biodegradation and tissue response of D,L-lactic acid oligomer microspheres of sizes less than 20 um, 20-100 pm and 100-200 urn injected into the knee joints of rabbits. Cell infiltration into the joint increased 4 fold for all microsphere size ranges compared to saline control animals, up to 9 h following intra-articular injection and then decreased to basal levels at 3 days.  Decreasing the  amount of microspheres injected into the joint from 20 mg to 5 mg per knee decreased the inflammatory response significantly (Nishide et al, 1999). The injected microspheres were found to be located primarily in the popliteal region of the knee joint.  Horisawa et  al (2002a) showed that both P L G A microspheres (26 um) and P L G A nanospheres (265 nm) were compatible and well tolerated in the joints of rats following intra-articular injections.  The microspheres were not phagocytosed and were observed to cause the  formation of granulation tissues surrounded by multinuclear giant-cells in the synovial  126  membrane.  The nanospheres, on the other hand, were extensively phagocytosed and  infiltrated through the synovial tissues inducing the proliferation of the synovium. In summary, 10% M T X loaded PLLA2k microspheres in the size range of 33-110 pm were developed and characterized as the lead formulation for further in vivo evaluation. The lifetime of M T X release and degradation/erosion of the microspheres was approximately 3-4 weeks.  127  Chapter 3 PHARMACOKINETICS OF METHOTREXATE FOLLOWING INTRA-ARTICULAR INJECTION OF METHOTREXATE LOADED MICROSPHERES 2  3.1  INTRODUCTION  M T X for the treatment of rheumatoid arthritis was introduced in the 1980's, and numerous studies have been conducted to evaluate the pharmacokinetics and pharmacodynamics of oral low dose M T X in patients with arthritis (Bannwarth et al, 1996; Grim et al., 2003). In human subjects, M T X is absorbed through the proximal jejunum, and about 5-7% of the administered M T X is metabolized by hepatic methotrexate 7-hydroxylase to 7-hydroxymethotrexate (7-OH-MTX) (Grim et al, 2003). Renal excretion accounts for between 60-90% of the M T X and 7-OH-MTX eliminated (Bannwarth et al., 1996). The antifolate activity of 7-OH-MTX is significantly less than M T X , but it is slowly excreted and may crystallize in the kidneys resulting in nephrotoxicity (Chen and Chiou, 1983). The hepatic methotrexate 7-hydroxylase activity in animals and humans has been shown to be quite variable, with highest activity  2  A version of this chapter has been published. Liang LS et al., (2005) Pharmacokinetic study of  methotrexate following intra-articular injection of methotrexate loaded poly(L-lactic acid) microspheres in rabbits. J Pharm Sci. 94(6):1204-15. 128  in rabbits, followed by rats, hamsters, and monkeys and undetectable in dogs (Chen and Chiou, 1983; Ivener al, 1985; Kitamura et al, 1999). For the evaluation of the pharmacokinetics of various drug delivery systems of M T X , rabbits are often chosen as the animal model for the reason that they have similar metabolic pathways to M T X to human subjects (Chen and Chiou 1982; Foong and Green, 1988; Wang et al, 1995). The pharmacokinetics of M T X following intravenous dosing ranging from 1 to 12 mg/kg in rabbits could be fitted to a linear three-compartment model with a terminal half-life between 2.4 to 3.6 h. For 8 h post-dosing, 50% of the dose of M T X was excreted into urine in the unchanged form and 15% as the metabolite 7-OH-MTX and these fractions were not influenced by changes in dose (Iven et al, 1985). The renal clearance of 7-OH-MTX was similar to M T X (Chen and Chiou, 1983). Renal clearance decreased with the increasing plasma levels, suggesting active tubular secretion as one of the excretion mechanisms (Iven et al, 1985; Chen and Chiou, 1983). Infusion studies of M T X and 7-OH-MTX in rabbits revealed that the metabolite has a longer residence time and a larger volume of distribution compared to M T X (Chen and Chiou, 1983). The pharmacokinetics of M T X following intra-articular injection of M T X solution and M T X encapsulated liposomal formulations have been investigated in rabbits (Foong  129  and Green, 1988). The data showed that M T X solution was rapidly cleared from the joint cavity and was detectable in plasma within 1 h of injection and 79% of the injected M T X dose was excreted in the urine within 24 h of injection.  The M T X encapsulated  liposomes, on the other hand, slowed down the clearance of M T X from joint cavity and 45% of the injected dose was recovered from the joint 24 h following injection (Foong and Green, 1988). M T X has not been previously formulated in polymeric microspheres for intra-articular injection and very few studies have been carried out to study the pharmacokinetics of a controlled release microspheres drug delivery system for intra-articular drug delivery. A knowledge of the pharmacokinetics and biodistribution of M T X following intra-articular injection of M T X loaded microspheres is considered fundamental to future in vivo efficacy studies. In this chapter, we report the results of two pharmacokinetic studies of M T X following intra-articular injection of M T X loaded PLLA2k microspheres (33-110 pm) into the knee joints of healthy rabbits.  The size range of 33-110 pm was selected based  on previous data in Chapter 1 showing good tolerability and biocompatibility in rabbit joints.  This size range of microspheres would likely remain trapped within the synovial  fluid since they are too large to be phagocytosed (Greis et al., 1994). Thus, M T X released from microspheres should be available for delivery to synovial tissues. M T X  130  release studies from M T X loaded PLLA2k microspheres showed that release of M T X was 95% complete in 2 weeks. Thus M T X loaded P L L A 2 k microspheres were felt to be an appropriate formulation for pharmacokinetic evaluation. The first study was a pilot study in which M T X solution or M T X loaded microspheres (dose of 1.5mg M T X ) was injected into the knee joints of the rabbits. The concentrations of M T X in plasma, urine and synovial tissues were determined.  The study was repeated with a higher dose of  M T X (lOmg) and histological analysis was also conducted in some of the rabbits. The objectives of these studies were: 1. To determine the plasma and urine concentrations of M T X and 7-OH-MTX following intra-articular injection of M T X loaded microspheres or M T X solution; 2.  To investigate the biodistribution of M T X following intra-articular injection of M T X  loaded microspheres or M T X solution; 3.  To determine the tissue locolization of M T X microspheres following intra-articular  injection of M T X loaded microspheres.  131  3.2 3.2.1  EXPERIMENTAL Materials  Methotrexate (MTX) (MW: 454.4g/mole) was purchased from Hande Tech Development Co. (U.S.A).  Aminopterin was purchased from Sigma Chemical Co. Poly  (vinyl alcohol) (PVA) (98% hydrated, M W : 13,000-23,000) was obtained from Aldrich Chemical Company Inc. 7-hydroxy-methotrexate (7-OH-MTX) was purchased from Schircks Laboratories (Jona, Switzerland). Poly(L-lactic acid) (MW: 1600-2400 g/mol, intrinsic viscosity: 0.1-0.2, polydispersity: 2-3) was obtained from Polysciences (Warrington, PA). (HPLC) grade.  A l l solvents used were High Performance Liquid Chromatography  Phosphate buffered saline (PBS, pH 7.4 ) was prepared by dissolving  0.32 g sodium dihydrogen orthophosphate, 2.15 g sodium phosphate, 8.22 g NaCl in one liter of distilled water.  Phosphate buffer (pH 6.5) was prepared by dissolving 0.936 g  sodium dihydrogen orthophosphate, 0.644 g sodium phosphate in one liter of distilled water and filtered through 0.2 mm filter paper.  3.2.2  Preparation of M T X loaded microspheres  The microspheres were prepared by the solvent evaporation method as described in Chapter 2.  Briefly, M T X was suspended in 20% PLLA2k polymer in methylene  chloride. The drug suspension was then slowly dispersed into 100 mL of a 2.5% P V A 132  solution and stirred at 1000 rpm using an overhead stirrer (BDC2002 Caframo, Ont, Canada). The resulting emulsion was continuously stirred for 2.5 hours at room temperature under ambient pressure until all the methylene chloride had evaporated. The solidified microspheres were recovered by centrifugation. The microspheres were sieved through 33 and 110 pm sieves and washed with distilled water.  The washed  microspheres were then air-dried overnight and stored in a desiccator at room temperature for further drying.  Either control or M T X loaded microspheres batches  intended for injection into rabbit joints were sterilized by gamma irradiation from a Co-60 source at a dose of 25kGy (Nordion International Inc.).  3.2.3  M T X and 7 - O H - M T X assays  3.2.3.1  MTX and 7-OH-MTX  extraction  M T X and the metabolite 7-OH-MTX in the plasma and urine samples were extracted according to the method reported by Cociglio et al (1995) with modifications. One mL acetonitrile was added to the plasma or urine sample (0.5 mL), vortex mixed and centrifuged at 16000 x g for 5 min. The supernatant was transferred to a 12 mL glass tube and 2 mL of methylene chloride were added to extract the aqueous phase. aqueous phase (200 pX) was dried under a stream of nitrogen gas at 45°C. sample was reconstituted with 200 pL phosphate buffer pH 6.5.  133  The  The dried  Five gram samples of organs and the whole joint tissues were homogenized in 5 mL PBS (pH 7.4) (Polytron, Brinkman Instruments, Mississauga, ON. Canada) and the homogenates were centrifuged at 16000 x g for 30 min.  The supernatants were removed  and processed in the same manner as the plasma and urine samples to extract M T X and 7-OH-MTX. To determine whether M T X plasma samples were stable during storage at -20°C, blank plasma and tissues samples (n=3) were spiked with 0.1 pg and 1 pg/mL M T X and stored at -20°C for 18 days.  The samples were then processed and analysed for M T X as  previously described. 3.2.3.2  HPLC assay for plasma and tissue samples  M T X and 7-OH-MTX in plasma and tissue samples were assayed by H P L C with fluorescence detection using the method developed by Albertioni et al (1995) with modifications.  The H P L C system consisted of a Shimadzu LC-10AD pump (Shimadzu  Corporation, Japan) and a Shimadzu SIL-9A automatic injector.  A Waters 470 scanning  fluorescence detector was used (Ex: 370nm, Em: 417 nm). A photochemical reactor unit with a 254 nm U V lamp and a 10 m x 0.25mm reaction coil (Aura Industries Inc, N Y , USA) was used post-column (before the detector).  The analytical column was a  Nov-Pak C18 column with dimensions of 3.9 x 150 mm. The mobile phase consisted of  134  10 m M phosphate buffer p H 6.5, with 3.2% acetonitrile and 6%> hydrogen peroxide (Sigma Chemical Co.) at a flow rate of lmL/min under ambient temperature.  The  injection volume was 50 pL. A folic acid antagonist, aminopterin (Figure 3.1), at a concentration of 0.2 pg/mL was used as the internal standard.  Blank plasma or tissue  supernatants were spiked with known concentrations of M T X and 7-OH-MTX and the internal standard.  Calibration curves were constructed by plotting the ratio of peak areas  of MTX/internal standard and 7-OH-MTX/internal standard versus concentrations of M T X and 7-OH-MTX, respectively. 3.2.3.3  HPLC assay for urine samples  M T X and 7-OH-MTX in the urine samples were assayed by H P L C with a U V detector.  The H P L C system consisted of a Waters 600S controller pump, a Waters 717  plus autosampler, and a Waters 486 absorbance detector (304 nm). The analytical column was a Novo-Pak C18 column with dimensions of 3.9 x 150mm. The mobile phase consisted of phosphate buffer pH 6.5, with 3.7% acetonitrile. The injection volume was 50 uL. Standard curves were constructed by plotting the peak area of M T X or 7-OH-MTX versus concentrations of M T X and 7-OH-MTX. was used in this assay.  135  No internal standard  3.2.3.4  Recovery of MTX and 7-OH-MTXfrom tissues  Two New Zealand White rabbits were sacrificed and their livers, lungs, spleens, kidneys and synovial tissues were removed and stored at -20°C until analysis. One gram samples of each tissue were spiked with 0.2 pg, 0.4 ug, and 4 pg M T X .  Four mL  of PBS ( pH.7.4) were added to the tissues and the tissues were processed the same way as described in section 3.2.3.1. The M T X recovery was measured by comparing the peak areas of M T X obtained from the tissue samples and the peak areas of M T X obtained from direct injections of 0.05, 0.1 and 1 pg/mL in phosphate buffer p H 6.5. 3.2.3.5  Validation of the assays  The assays for M T X in plasma and urine were validated by measuring four sets of standards on three separate days. were characterized.  The retention time and specificity of the M T X peak  The linearity, limit of detection, limit of quantitation, inter-day  precision, intra-day precision, and accuracy of the calibration curve were measured. The specificity of the assay was determined by observing blank replicates for lack of peaks at the retention time of M T X .  Linearity was expressed as the coefficient of  determination (R ) for each of the twelve calibration curves assembled over three days 2  and for the standard curve consisting of each of the twelve standards' data points for each concentration.  The limit of detection was calculated from Equation 2.1 and the limit of  quantitation was calculated from Equation 2.2 (given in Chapter 2). 136  Inter-day precision was expressed in terms of the coefficient of variation (CV) of the mean of the twelve standards' data points collected over three days for each concentration.  Intra-day precision was expressed in terms of the C V of the mean of the  four standards' data points on each day for each concentration. Accuracy was determined by assembling a calibration curve and comparing three different standards at each concentration against the calibration curve each day for three days. Accuracy was expressed in terms of the percent deviation (bias) of each of the three standards and their mean bias compared to the daily calibration curve.  3.2.4  Microspheres recovery from joint fluids  Two New Zealand White rabbits were injected with 56mg of 18% loaded microspheres in the right hind knee joint while under inhalation anesthesia. were immediately sacrificed.  The rabbits  One mL of heparinized (lOunits/mL) saline was injected  into the right knee joint of each rabbit following sacrifice of the rabbits. The right knee joints of the rabbits were flexed and massaged and then the fluid in the joints was aspirated, with a syringe with a 23 gauge needle.  The washing and aspiration of the right  knee joint was repeated with another l m L of heparinized saline. The aspirated fluids were pooled and centrifuged at 325 x g. The supernatant was removed and analysed for M T X by UV-vis spectrophotometry.  The sedimented microspheres were dissolved in  137  l m L of methylene chloride and M T X encapsulated in the microspheres was extracted with l m L of PBS (pH 7.4) and analysed by UV-vis spectrophotometry.  3.2.5  Pharmacokinetic studies in rabbits  Both pharmacokinetics studies were conducted according to the protocol # A02-0066 approved by the U B C animal care committee.  The experiments were  conducted at the Animal Care Facility of the Vancouver Hospital and Health Sciences Center.  The animals were housed in metal cages and monitored by Mr. Michael Boyd.  The room was maintained at a temperature of approximately 20°C with 30-70% relative humidity and a light/dark cycle of 12 hours/12 hours.  Rabbit chow and tap water were  provided ad libitum to the animals for the duration of the study. 3.2.5.1  Low dose pharmacokinetic study  Fourteen female New Zealand White Rabbits (weight 2.5-3.0 kg) were surgically implanted with catheters (Access Technologies Inc.) into the jugular veins.  The  animals were randomized to receive an intra-articular injection into the right knee joint cavity of either 1.5 mg M T X solution in 200 uL PBS ( M T X solution) or 25 mg of 6% M T X loaded P L L A 2 k microspheres (sized 33-110 pm) in 200 uL PBS.  The M T X  solution was prepared by dissolving 15mg M T X in 500 pL sodium hydroxide (0.4N) followed by dilution with PBS to 2mL. The pH remained at 7.0.  138  Total dose of M T X in  the microspheres was equivalent to 1.5 mg. Serial blood samples (1.5 mL) were obtained from the jugular vein immediately before intra-articular injection and at 5, 15, and 30 minutes, 1, 2, 4, 6, 8, 10, 24, and 48 hours after the intra-articular injection. A n equivalent volume of 0.9% sodium chloride injection was administered into the animal after each blood draw.  The blood samples were centrifuged for 10 minutes at 16000 x g,  and the plasma harvested and stored at -20°C before drug analysis. Urine samples of the rabbits were collected from the trays below the cage of the rabbits at 0-8 h, 8-24 h, and 24-48 h intervals. The animals were sacrificed 48 hours following the intra-articular injection and major organs (livers, kidneys, lungs, heart, and spleen) and the joint tissues were removed from the rabbits for further analysis. 3.2.5.2  High dose pharmacokinetic study  Sixteen female New Zealand White Rabbits (weight 2.5-3.0 kg) were used for the study. The animals were randomized to receive an intra-articular injection into the right knee joint cavity of either 10 mg M T X solution in 400 uL PBS or 56 mg of 18% M T X loaded PLLA2k microspheres (sized 33-110 um) in 400 uL PBS.  The M T X stock  solution was prepared by dissolving 100 mg M T X in 500 uL sodium hydroxide (0.4 N) followed by dilution with PBS to 4 mL. The pH remained at pH 7.0.  Total dose of M T X  in the microspheres was equivalent to 10 mg. Immediately before (0 h) and following  139  the intra-articular injection, serial blood samples (1.5 mL) were obtained via the jugular vein at 5, 15, and 30 min, 1, 2, 3, 4, 5, 6, 8, 24 h after the intra-articular injection. A n equivalent volume of normal saline was administered into the animal after each blood draw. The blood samples were centrifuged for 10 min at 16000 x g, and the plasma harvested and stored at -20°C for drug analysis. To ensure complete collection of urine samples, the animals were kept under light inhalation halothane (1.5%) anesthesia and the urine samples were collected from a catheter inserted into the urethra of the rabbits in the first 8 h following intra-articular injection. Between 8 to 24 h, the urine samples were collected from the trays below the cage of the rabbits.  Eight rabbits (four injected  with M T X solution, four injected with M T X microspheres) were sacrificed 6 h following intra-articular injection, and the rest of the rabbits were sacrificed 24 h following intra-articular injection. The synovial joint was flushed with 2 mL of heparinized saline and the fluid was aspirated for drug analysis. Major organs (livers, kidneys, lungs, heart and spleen) and the joint tissues were removed from the rabbits for further analysis.  3.2.6  Pharmacokinetic calculations  The plasma curves were resolved with the nonlinear regression computer program WinNONLIN (Scientific Consulting Inc., Standard Edition, and Version 1.1). terminal rate constant, and mean residence time were calculated using the  140  The  non-compartmental analysis with extravascular dose input. The 1 order rate constant st  (k) associated with the terminal portion of the M T X plasma curve was estimated via linear regression of time versus log concentration. The elimination half-life was determined by: uEquation •• * 3.11  t, 1 = ° k  6 9 3  2  The area under the concentration-time curve (AUCo-oo) and area under the moment curve was calculated by the trapezoidal rule for the observed values and then extrapolated to infinity as shown in Equations 3.2 and 3.3. A UCo-oo = A UCo- + A UC,.00 = £ t  C +C  N  AUMCo- =  X  x  n=l  OVi ~ n) f  C  "  + C  2 (t +t  " + ' (f„ -t +1  1  }  +  C -t  n)  + ^k C + 2  Equation 3.2 Equation3.3  where the summation is calculated over N trapezoids formed by n +1 data points. C and t represent the concentration and the time after administration, respectively. C is t  the plasma concentration of M T X at the last data point. The mean residence time (MRT) was calculated by MRT=  A  ° AUC _ U  M  Equation 3.4  C  0  X  The plasma data were also fitted with a one-compartment and a two-compartment model with first order dose input to determine the absorption constant and terminal  141  elimination constant.  The Akaike's information criterion suggested that the  one-compartmental model (AIC: -57) was more efficient in describing the plasma data compared to the two-compartmental model (AIC: -23). Therefore, the plasma data were fitted to a one-compartment model and the absorption rate constant was determined from equation 3.5. C(t)= Dok /V(k -ke)(e- - e' ) ket  a  Equation 3.5  kat  a  where Do and V represent dose administered and volume of distribution, respectively. k and k are absorption and elimination rate constants, respectively. The a  e  absorption half life ti/2(abs) was calculated by: ti/2(abs)= ®^®L K  3.2.7  Equation 3.6  Histological analysis  In the high dose study (10 mg M T X ) , one of the knee joints from both treatment groups, sacrificed at 6 h and 24 h following intra-articular injection, were used for histological analysis. for 7 days.  The knee joints were removed in toto and fixed in 10% formalin  The muscles were trimmed off and the knee joints were fixed in  decalcification fluid, which consisted of 10% formic acid in 4% formaldehyde, for 4 weeks.  Sections were cut in the 8 sagittal planes and stained with haematoxylin/eosin.  142  3.2.8  AST analysis of rabbit plasma samples  The asparate aminotransferase (AST) activities in the rabbit plasma samples from the high dose pharmacokinetic study were analysed using an A S T transaminase analysis kit (Sigma Diagnostics). The principle of AST analysis is based on the colorimetric method proposed by Reitman and Franke (1957).  The reaction of aspartic acid and  a-ketoglutaric acid is catalyzed by AST to form oxalacetic acid and glutamic acid. The oxalacetic acid is then reacted with a color reagent, 2,4-dinitrophenylhydrazine to form phenylhydrazones.  The color intensity of phenylhydrazones is proportional to the  transaminase activity. A set of standards was first prepared using the transaminase calibration standard solution (sodium pyruvate) and substrate solution which contains DL-asparate, and a-ketoglutaric acid. A calibration curve of the U V absorbance values at the wavelength of 505 nm versus the corresponding units of AST (0-216 SF units/mL) was created. To analyse the activity of AST in plasma, 20 pL of plasma sample was added to the substrate solution (100 uL) and kept at 37°C for an hour to allow the reaction between the substrate solution and A S T in the plasma.  Color reagent containing  2,4-dinitrophenylhydrazine (100 uL) was then added and the whole mixture was shaken gently and left at room temperature for 20 min to allow the reaction between oxalacetic  143  acid and 2,4-dinitrophenylhydrazine.  One mL of 0.40 N Sodium Hydroxide Solution  was then added to the mixture to terminate the reaction. The U V absorbance of the mixture was then read and recorded at 505 nm using a U V - V i s spectrophotometer with water as the reference.  A S T activity was determined in Sigma-Frankel (SF) units/mL  from the calibration curve.  3.2.9  Statistical analysis  Student's t test (p<0.05) was used to compare the concentrations of M T X in various samples and pharmacokinetic parameters between two treatments.  144  NH  2  /  V<  )s-s^ r \ / \ = C H \ //> — C ;\ 7 10  /•  =c  CH—CH  / CH -CH 2  NH-HC  \  //  Figure 3.1  Chemical structure of aminopterin.  145  0  C  OH  2  3.3  RESULTS  3.3.1  M T X and 7 - O H - M T X assay validation  3.3.1.1 Specificity, stability, and recovery M T X and 7-OH-MTX in plasma samples and tissue samples were analyzed by HPLC using a post column U V reactor and a fluorescence detector.  The assay was able  to separate internal standard aminopterin, M T X and 7-OH-MTX into three distinctive peaks with retention times of 7.3, 19.8 and 25.1, minutes respectively.  Blank rabbit  tissue samples gave chromatograms that did not exhibit any peaks at these positions. Attempts were made to analyze M T X in urine using the H P L C assay with the fluorescence detection. The substances in the urine produced significant peaks around the peak positions of M T X and 7-OH-MTX.  Therefore, an H P L C assay with a U V  detector was developed. The assay was able to separate M T X and 7-OH-MTX into two distinct peaks with retention times of 11.9 and 15.2 minutes, respectively. Blank urine samples showed no peaks at positions for M T X and 7-OH-MTX but peaks were present in the position for aminopterin. Therefore, no internal standard was used in this assay. The rabbit plasma and tissue samples were spiked with 0.05, 0.1 and 1 pg /mL MTX.  A sample of plasma and each tissue was also spiked with 0.1 pg and 1 pg /mL  M T X and stored at -20°C for 18 days. The percent recovery of M T X from these tissues  146  was calculated by extraction and H P L C analysis of these samples and comparing the M T X peak area obtained with those from control assays, in which M T X was present but no tissue was used.  Table 3.1 shows the percent recovery data for each tissue, plasma  and urine tested at different M T X concentrations.  The percent recovery of M T X from  liver, kidneys and synovial tissues was approximately 50% at 3 different concentrations. The recoveries of M T X from plasma and urine were over 90% at 1 and 0.1 pg/mL, but dropped to 70% and 85% for plasma and urine samples, respectively, when spiked with 0.05pg/mL of M T X .  Recovery of M T X from plasma and tissue samples spiked at 0.1 pg  and lug and stored at -20°C over 18 days did not show any significant differences when compared with the recovery of M T X from tissues processed immediately, suggesting that M T X was stable in the rabbit tissues stored at -20°C over this period of time. 3.3.1.2 Precision, accuracy, linearity, range, and limit of detection The assays for M T X and 7-OH-MTX in plasma and urine were validated over three days.  A linear relationship between M T X or 7-OH-MTX concentrations in plasma  and ratios of peak areas of M T X or 7-OH-MTX to the internal standard were obtained in the concentration range of 0.01 to lug/mL (R >0.99) by the H P L C assay. 2  Representative standard curves in the concentration range of 0.01 to 1 pg/mL M T X and 7-OH-MTX in plasma are shown in Figure 3.2. The 3-day inter and intra-day precision  147  data are given in Table 3.2. was about 14%.  The overall inter-day and intra-day coefficient of variation  Acceptable assay precision" is generally considered to be < 20% at the  lowest concentration measured and < 15% at all other concentrations for the % C V values.  These criteria were met using this H P L C assay at concentration range of 0.01  pg/mL to 1 pg/mL for M T X . The limit of detection of M T X and 7-OH-MTX in plasma were 0.038 pg/mL and 0.046 pg/mL respectively, and the limit of quantitation was 0.13 pg/mL for M T X and 0.14 pg/mL for M T X and 7-OH-MTX, respectively. The accuracy data are summarized in Table 3.3. A C V value less than 15% at concentration range from 0.05 to 1 pg/mL indicates sufficient accuracy at this range. For analysis of M T X and 7-OH-MTX in urine samples, the overall inter-day and intra-day coefficient of variation was less than 15% in the concentration range of 0.5 to 20 ug/mL (Table 3.4).  A linear relationship between M T X concentrations in urine and  peak areas of M T X was obtained in the concentration range of 0.1 to 20pg/mL (TC>0.99) by the H P L C assay with U V detection (Figure 3.3).  The limit of detection of M T X and  7-OH-MTX in plasma were 0.13 p-g/mL and 0.14 pg/mL respectively, and the limit of quantitation was 0.4 ug/mL for M T X and 0.43 ug/mL for M T X and 7-OH-MTX, respectively.  148  Table 3.1  Percentage of M T X recovered from tissues, plasma and urine samples spiked  with known concentrations of M T X following extraction and H P L C analysis. The values are shown as the mean of 3 samples ± one standard deviation (n=3).  MTX recovered (%) " Spiked M T X  Synovial  concentration  tissues  Kidneys  Liver  Plasma  Urine  (pg/mL) 0.05  57.5±1.5  58.1+1.5  58.0±2.5  70.0±0.7  85.1+0.5  0.1  49.8±0.8  60.0±2.7  59.2+1.4  95.1+1.6  90.4+0.8  1  50.1±2.9  48.1±3.9  60.1±0.6  94.3+0.8  95.5+1.4  a  M T X recovered was measured by dividing the peak areas of M T X obtained from the  tissue, plasma and urine samples by the peak areas of M T X obtained from direct injections of 0.05, 0.1 and 1 pg/mL in phosphate buffer pH 6.5 and expressed as a percentage.  149  A)  MTX in plasma (p.g/mL)  0  0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8  0.9  1  1.1  7-OH-MTX in plasma (ug/mL)  Figure 3.2 Representative standard curves of A) M T X and B) 7-OH-MTX extracted from rabbit plasma and analysed by H P L C . Peak area ratio is the ratio of peak area of M T X or 7-OH-MTX to the peak area of aminopterin measured from H P L C chromatographs.  R is the coefficient of determination. 2  150  A) y= 104334x+ 173067  (3  PH  10 B)  15  25  20  M T X in urine (ug/mL)  2.5 y=84181x+91044 O  x CO  10  15  25  20  7-OH-MTX in urine (ug/mL)  Figure 3.3  Representative standard curves of A) M T X and B) 7-OH-MTX extracted  from rabbit urine and analysed by H P L C . Peak area is the area under the peak of either M T X or 7-OH-MTX measured from H P L C chromatographs. R is coefficient of the 2  determination.  151  Table 3.2  Inter- and intra-day precision for the H P L C assay with fluorescence  detection for A) M T X , B) 7-OH-MTX extracted from rabbit plasma. A) MTX concentration in plasma (Ug/mL)  Day 2  Dayl %CV  Peak  b  ratio  Peak  %CV  Peak  b  area  area  ratio  3  All days  Day 3 %CV  b  ratio  b  area  area 3  %CV  Peak ratio  3  3  0.005  0.0038  35.5  0.0033  42.5  0.0015  52.6  0.0035  52.3  0.01  0.0044  17.4  0.0042  19.5  0.0064  18.8  0.0045  16.9  0.05  0.0267  9.9  0.0336  6.6  0.0325  11.5  0.0309  12.0  0.1  0.0511  5.8  0.0571  14.1  0.0685  12.7  0.0589  14.5  0.5  0.2696  5.1  0.2627  6.5  0.2842  7.2  0.2720  10.1  1  0.5362  6.0  0.5738  4.5  0.6924  2.7  0.6008  13.6  B) 7-OH-MTX concentration in plasma (Mg/mL)  %cv  Peak  b  ratio  Peak  %CV  b  ratio  3  Peak  %CV  b  ratio  3  Peak  %CV  b  area  area  area  area  All days  Day 3  Day 2  Day 1  ratio  3  3  0.005  0.0035  34.5  0.0043  45.1  0.0014  32.1  0.0021  41.5  0.01  0.0049  17.8  0.0048  13.3  0.0054  24.5  0.0048  20.1  0.05  0.0349  9.9  0.0352  14.5  0.0360  6.9  0.0342  15.4  0.1  0.0698  7.5  0.0682  5.8  0.0722  1.3  0.0752  13.1  0.5  0.3477  5.1  0.3172  6.3  0.3641  6.1  0.3430  6.9  1  0.6520  8.2  0.6677  10.9  0.7484  2.5  0.7227  5.1  3  Peak area ratio is the ratio of peak area of M T X or 7-OH-MTX to the peak area of  aminopterine. The value reported is the average calculated from four standard curves at each concentration for each day (n=16 for all days). b  % C V is the coefficient of variation, which is the ratio of the standard deviation to the  mean area ratio values, expressed as a percentage. A value less than 20% at the lowest concentration and less than 15% at all other concentrations was taken to indicate sufficient precision.  152  Table 3 . 3  Accuracy for the H P L C assay with fluorescence detection for A) M T X , B)  7-OH-MTX extracted from rabbit plasma. A) MTX  Dayl  concentration  Measured  (fig/mL)  value  Day3  Day 2  %cv  b  Bias  0  (%)  3  Measured  %CV  value  Bias  Measured  (%)  value  %CV  Bias  (%)  0.01  0.013  39.5  30.0  0.012  14.5  20.0  0.015  34.2  50.0  0.05  0.058  8.4  16.0  0.067  5.7  34.30  0.056  3.1  12.0  0.1  0.103  5.4  3.1  0.10  15.7  4.74  0.11  14.5  9.7  0.5  0.51  6.5  1.3  0.47  8.9  -6.86  0.45  10.8  -9.9  1  0.98  8.3  -1.6  1.03  1.5  3.43  1.05  2.4  5.3  B) 7-OHMTX  Dayl  concentration  Measured  (jug/mL)  value  3  Day3  Day 2  %CV  b  3  Bias  Measured  (%)  value  %CV  3  Bias  Measured  (%)  value  %CV  b  0.01  0.015  34.8  50.0  0.015  45.8  50.0  0.014  29.4  40.1  0.05  0.061  7.5  22.0  0.057  15.4  14.0  0.053  10.1  6.0  0.1  0.101  5.5  1.1  0.114  4.8  14.1  0.099  4.7  -1.3  0.5  0.541  5.9  7.1  0.510  7.7  2.0  0.432  4.1  -13.7  1  0.998  9.9  -1.5  1.068  12.7  6.8  1.058  1.9  2.8  Measured value is the average (n=3 for each day) of values of concentration calculated  from one standard curve on that day. b  Bias  (%)  3  % C V is the ratio of the standard deviation to the average, expressed as a percentage. A  value less than 15% indicates sufficient accuracy at each concentration. °Bias is the ratio of the deviation of measured value from the actual concentration measured, expressed as a percentage.  153  Table 3.4  Inter- and Intra-day precision of the H P L C assay with U V detection for A)  M T X , B) 7-OH-MTX extracted from rabbit urine. A) MTX concentration (fig/mL)  Day 2  Dayl  Peak Area  %CV  b  Day 3  Peak  Peak  %CV  %CV  Peak  %CV  Area  Area  Area  3  All days  0.1  35457  4.8  38132  20.5  301445  31.4  33785  31.6  0.5  72657  16.5  67887  13.4  61292  10.7  63945  17.4  1  154065  6.4  146987  8.7  145478  4.6  154785  6.4  5  687121  4.7  727697  11.6  702109  2.9  672309  5.4  10  1454386  8.5  1528885  10.4  1516005  9.8  1512974  10.4  20  2847956  9.4  2757286  4.7  2920613  5.7  2733949  8.8  B) 7-OHMTX concentration (fig/mL)  Peak Area  Day 3  Day 2  Dayl  3  %  Peak  CV  Area  %CV  Peak  All days  %CV  Peak  %CV  Area  Area  0.1  38233  20.4  29878  15.7  20457  25.4  59365  20.3  0.5  82767  13.9  98787  17.3  87898  16.8  97086  19.1  1  178329  5.8  180428  4.4  153483  10.8  162412  10.6  5  751731  9.3  853864  6.3  742016  5.4  852848  12.5  10  1726327  8.3  1901803  9.1  1801428  2.3  1727183  13.1  20  3548789  5.4  3372828  5.8  3617751  3.7  3521411  8.5  a  Peak area ratio is peak area of M T X or 7-OH-MTX measured from H P L C  chromatographs.  The value reported is the average calculated from four standard curves  at each concentration for each day (n=16 for all days). b  % C V is the coefficient of variation, which is the ratio of the standard deviation to the  mean area ratio values, expressed as a percentage. A value less than 20% at the lowest concentration and less than 15% at all other concentrations indicated sufficient precision.  154  3.3.2  Microspheres recovery from rabbit joint  To investigate the feasibility of recovering M T X and M T X loaded microspheres from the synovial fluid of the rabbits, two rabbits were injected intra-articularly with 56 mg of 18 % M T X loaded microspheres, and the joints were immediately flushed with 2 mL of heparinized saline and the fluids in the joints were aspirated for analysis. The pooled fluids from synovial joints were clear in appearance but there was evidence of microspheres retrieved from the joint space via the needle and syringe. The amounts of M T X in the aspirated fluids and microspheres were analysed and the results are shown in Table 3.5. About 10 % of the injected dose was recovered from the joints of the rabbits with 3-4 % of the dose found in the aspirated fluids and 7-8 % of the dose was found in the retrieved microspheres.  155  Table 3.5  The amounts of M T X recovered from aspirated fluid from rabbit joints  immediately following intra-articular injection of 56 mg of 18% M T X loaded PLLA2k microspheres. Rabbit #  a  A mt of MTX in  Amt of MTX in  MTX  fluid aspirated thefluid(p.g)  microspheres  recovered (%)  (mL)  12.9 10.3  Volume of  1  1.5  437.6  (f*g) 856  2  1.2  304.6  725  0  M T X recovered was calculated by dividing the total amount of M T X determined in the  joint (amount in the fluid + amount in the microspheres) by the dose of M T X injected (lOmg), expressed as a percentage.  156  3.3.3  3.3.3.1  Pharmacokinetic studies in rabbits  Low dose pharmacokinetic study The plasma M T X concentrations following a single intra-articular injection of  either 25 mg of 6% (w/w) M T X loaded PLLA2k microspheres (33 -110 pm) or 1.5 mg M T X solution are shown in Figure 3.4. The plasma M T X concentration rose 5 min after the injection of M T X solution and plasma concentrations reached a maximum 15 min after the injection (t  max  of 15 min and C  m a x  of 0.6 pg/ml). The plasma concentrations  gradually declined and were undetectable 8 h after the injection. For the rabbits injected with M T X loaded microspheres, the plasma M T X concentrations plateaued at about 0.06 pg/ml between 5 min and 3 h and were undetectable, 4 h following the injection.  The  concentrations of the major metabolite 7-OH-MTX in plasma were not analysed since the 7-OH-MTX standard was not available at the time this study was conducted. The amounts of M T X excreted in the urine of rabbits following the intra-articular injections of microspheres and M T X solution are shown in Figure 3.5.  Some samples of  urine were missed during the time period of 0-3 h as the rabbits were recovering from anesthesia in a heated incubator following intra-articular injection. The results showed that in the period of 8 to 24 h post intra-articular injection, the amount of M T X excreted into the urine from rabbits injected with M T X solution was more than 6 fold higher than  157  that from rabbits injected with M T X loaded microspheres and the difference was statistically significant (p < 0.05). In the period of 24-48 h, similar amounts of M T X were excreted from both groups. The total amount (0-48 h) of M T X excreted into the urine from rabbits injected with M T X solution was also significantly higher (p < 0.05) than that from rabbits injected with M T X loaded microspheres. There was no M T X detected in the major organs from either rabbits injected with M T X solution or M T X loaded microspheres.  When analyzing synovial tissues,  approximately 0.6 pg of M T X was detected in the joint tissues of two of the rabbits injected with M T X microspheres while no M T X was detected in the joints of rabbits injected with M T X solution.  158  .S  1  -i  0  1  2  1  3  1  1  1  4  5  6  1  7  1  8  1  9  Time (h)  Figure 3.4  M T X concentrations in rabbit plasma following a single intra-articular  injection of either M T X solution or 25mg of M T X loaded P L L A 2 k microspheres (33-110 pm) in 200 uL PBS. Values are mean ± one standard deviation, (n=7). The dose of M T X injected was 1.5mg. * Indicates statistical difference between M T X plasma concentrations of rabbits injected with M T X solution and M T X loaded microspheres by paired t-test (p<0.05).  ( • ) M T X solution, (•) M T X microspheres.  159  300 • M T X solution 00 3.  J3 T3  250  H M T X microspheres  200 150  o  X  H  100 50 0 0-8 h  8-24 h  24-48 h  total 0-48 h  Time of urine collection  Figure 3.5  The amount of M T X (pg) excreted at different time periods in the urine of  rabbits following a single intra-articular injection of either M T X solution or 25mg of M T X loaded PLLA2k microspheres (33-110 pm) in 200uL PBS. * Indicates statistical difference between the amounts of M T X excreted from rabbits injected with M T X solution and M T X loaded microspheres by paired t-test (p<0.05). "Part of the urine sample was missed from both groups during this time period. Values are mean ± one standard deviation, (n=3). The dose of M T X injected was 1.5mg.  160  3.3.3.2  High dose pharmacokinetic study  The plasma M T X concentrations following a single intra-articular injection of either 10 mg M T X solution in 400 u.L PBS or 56 mg of 18% M T X loaded PLLA2k microspheres (33-110 pm) in 400 pL PBS are shown in Figure 3.6.  Plasma M T X  concentrations rose 5 min following intra-articular injection of M T X solution, and peaked 15 min following injection  (t ax m  of 15 min, C  m  ax  of 2.64 pg/mL).  The  concentration gradually declined and was still detectable 8 h following injection. For rabbits injected with 56 mg of 18% M T X loaded microspheres, the plasma M T X was detectable 5 min following injection and peaked 15 min following injection min,  C ax m  of 0.4 pg/mL).  (t x ma  of 15  The concentration gradually declined and was below the  detection limit 6 h following injection. The plasma concentrations of 7-OH-MTX, the major metabolite of M T X , following the injection of M T X solution are shown in Figure 3.7.  The data for 2 animals (rabbit #  5 and rabbit #15) are shown separately from the rest of the group due to significant inter-individual variability.  The metabolite 7-OH-MTX was detectable in the plasma of  rabbits injected with M T X solution 5 min following injection.  With the exception of  two rabbits (# 5 and #15), the plasma concentrations of 7-OH-MTX following the injection of M T X solution peaked at 1 h (Cmax 1.19 ug/mL) and gradually declined to  161  undetectable levels 6 h following the injection.  The concentrations of 7-OH-MTX  were higher than M T X in plasma from 2 h to 8 h post injection.  The peak 7-OH-MTX  concentrations in rabbit #5 and #15 occurred at 4 h following injection and were 3 to 5 fold higher than the peak concentration of the rest of the group. For rabbits injected with M T X microspheres, the concentration of 7-OH-MTX peaked at 1 h following injection (Cmax 0.2 pg/mL) except for two rabbits (rabbit # 4 and rabbit #14) (Figure 3.8). The peak 7-OH-MTX concentrations for rabbit #4 and rabbit #14 occurred at 2 h following injection and were 3 to 8 fold higher than the mean peak concentration for the rest of the group. The amounts of M T X (pg) excreted at different time periods in the urine of rabbits after a single intra-articular injection of either M T X solution or M T X loaded microspheres are shown in Figure 3.9A.  Between 0 to 3 h, about 1500 pg of M T X was  excreted in to the urine from the rabbits injected with M T X solution and this accounted for 15% of the total dose of M T X injected.  From 3 to 6 h and from 6 to 24 h, an  additional 2.6% and 5.7% of the total dose were excreted, respectively. On the other hand, less than 2% of the total dose of M T X was excreted in the urine of those rabbits injected with 18% M T X loaded microspheres from 0 to 3 h and 3 to 6 h time periods and 2.48% of the total dose of M T X was excreted in the period of 6-24 h.  162  The amounts of 7-OH-MTX excreted in the urine following a single intra-articular injection of either M T X solution or M T X loaded microspheres are shown in Fig 3.9B. Between 0-3 h, about 1700 pg of 7-OH-MTX were excreted in the urine of the rabbits injected with M T X solution. Between 3-6 h and 6-24 h, an additional 800 and 1100 pg of 7-OH-MTX, respectively, were excreted in the urine.  For rabbits  injected with M T X microspheres, less than 200 pg of 7-OH-MTX were excreted in the period of 0-3 h and an additional 152 and 260 pg of 7-OH-MTX, respectively, were excreted in the urine between 3-6 h and 6-24 h, respectively.  In terms of the percentage  of total dose, approximately 37% of the total dose was excreted as 7-OH-MTX from rabbits injected with M T X solution while 5.7% of the total dose was excreted as 7-OH-MTX from rabbits injected with M T X microspheres. To evaluate whether drug loaded microspheres were retained in the joint 6 h and 24 h following intra-articular injection of microspheres, synovial fluid and synovial tissues of the rabbits were removed for analysis.  The joint was flushed with 2 mL heparinized  saline and the fluid in the joint was aspirated and analyzed for M T X . The amounts of M T X in the aspirated fluids 6 h and 24 h following a single intra-articular injection of either M T X solution or M T X loaded microspheres are given in Figure 3.1 OA.  At 6 h  following the intra-articular injection, the amount of M T X in aspirated fluid of the rabbits  163  injected with microspheres was about 9 fold higher than that of the rabbits injected with M T X solution.  At 24 h following the intra-articular injection, no M T X was detected in  the aspirated fluid of the rabbits injected with M T X solution while the amount of M T X in the aspirated fluid in the rabbits injected with M T X microspheres decreased to 0.02 ug. The synovial tissues of the rabbits were removed and analyzed for M T X after the animals were sacrificed. given in Figure 3.1 OB.  The amounts of M T X in the synovial tissues of the rabbits are Since the recovery of M T X from the synovial tissue samples  was 50% (Table 3.1), the results reported in Figure 3.1 OB were corrected to account for 100% recovery. At 6 h after intra-articular injection, the amount of M T X in the synovial tissues of the rabbits injected with M T X microspheres was about the same as in the synovial tissues of the rabbits injected with M T X solution.  At 24 h after intra-articular  injection, the amount of M T X in the synovial tissues of the rabbits injected with M T X microspheres was about three fold higher than that of the rabbits injected with M T X solution. Samples of liver and kidneys were analyzed for the concentration of M T X . No M T X was detected in liver tissues from either group of rabbits. Analysis of kidney samples demonstrated that only rabbits administered M T X solution and sacrificed at 6 h following injection showed M T X present in the kidneys at about 0.1 pg/5g tissue.  164  4  r  3.5 3 -  Time (h)  Figure 3.6  M T X concentrations in rabbit plasma after a single intra-articular injection  of either M T X solution or 56mg of 18% M T X loaded PLLA2k microspheres (33-110 pm) in 400 pL PBS.  Values are mean ± one standard deviation, (n = 8 for 0-6 h period,  n = 4 for 6-24 h period).  The dose of M T X injected was lOmg.  ""Indicates statistical  difference between M T X plasma concentrations of rabbits injected with M T X solution and M T X loaded microspheres by student's t-test (p<0.05). M T X microspheres.  165  (•) M T X solution, (•)  6 r  0  5  10  15  20  25  30  Time (h)  Figure 3.7  7-OH-MTX concentrations in rabbit plasma following a single  intra-articular injection of lOmg M T X solution in 400 pX PBS. (•) rabbit #15, (•) rabbit #5, ( A ) the remaining animals in the group, values are mean ± one standard deviation (n = 6).  166  0  2  6  4  8  10  Time (h)  Figure 3.8  7-OH-MTX concentrations in rabbit plasma following a single  intra-articular injection of 56 mg of 1 8 % M T X loaded PLLA2k microspheres (33-110 um) in 4 0 0 u.L PBS. The injections provided a 10 mg M T X dose.  (A)  rabbit # 1 4 , ( « )  rabbit # 4, ( • ) the remaining animals in the group, values are mean ± one standard deviation (n = 6)  167  A) 3500 r oo  3000 o  2500  c -o  2000  .g  0) o X  • M T X solution H M T X microspheres  1500  X H  1000 500 L...T".,..  0 0-3 h  3-6 h  6-24 h  total 0-24 h  Time of urine collection  B)  00 .3 <L>  J3 C  T3 U •*-» <U  o X <u  X  H  S I  K O  I  5000 4500  • M T X solution  4000  H M T X microspheres  3500 3000 2500 2000 1500 1000 500 0 0-3 h  3-6 h  6-24 h  total 0-24 h  Time of urine collection  Figure 3.9  The amount of A) M T X (pg) and B) 7-OH-MTX (pg) excreted at different  time periods in the urine of rabbits following a single intra-articular injection of either 10 mg M T X solution in 400 uL PBS or 56mg of 18% M T X loaded PLLA2k microspheres (33-110 pm) in 400 pL PBS. Values are mean ± one standard deviation, (n = 8 for the period of 0-6 h, n= 4 for the period of 6-24 h). The dose of M T X injected was lOmg. * Indicates statistically different between two treatments by stduent's Mest (p<0.05).  168  0.7 A)  'So 23.  -4->  Q o  r  • M T X solution  0.6 -  H M T X microspheres  0.5 0.4 -  so '3 ta  .5 X H  0.3 0.2 0.1 0 6h  24 h  Time following intra-articular injection  0.9  B)  0.8 00  • M T X solution M M T X microspheres  0.7  3  0.6 0.5  >  o  S3  0.4 X  0.3  H  0.2 0.1 0 24 h  6h  Time following intra-articular injection  Figure 3.10  The amount of M T X in A) pool fluid aspirated from synovial joints and B)  the synovial tissues, 6 h and 24 h following a single intra-articular injection of either 10 mg M T X solution in 400 uL PBS or 56mg of 18% M T X loaded microspheres in 400 uL PBS. The values reported in B) are corrected to account 100% M T X recovery from synovial tissues.  Values are mean ± one standard deviation, (n = 4). The dose of M T X  injected was lOmg. * Indicates statistically different between two treatments by student's t-test (p<0.05).  169  3.3.4  Pharmacokinetic analysis of plasma data  The plasma data of rabbits in both low dose and high dose pharmacokinetic studies were analysed using a non-compartmental and a one-compartmental model with extravascular dose input. The values of elimination half-life (ti/2) determined using both the non-compartmental and the one-compartmental models were similar and therefore, only the values that were determined by the non-compartmental model are reported. The pharmacokinetic parameters for rabbit plasma data from both experiments are given in Table 3.6. At both doses, the area under the curve (AUCo-oo) of the group injected with M T X solution was about 7 fold higher than the AUCo-oo of the group injected with M T X microspheres, while no significant difference was observed in elimination half-lives (ti/ ) and mean residence times (MRT) between the groups at either dose (Table 2  3.6). The absorption half-lives determined from one-compartmental analysis varied from 0.04 h to 0.13 h (Table 3.6) but were not significantly different at either dose.  170  Table 3.6 Pharmacokinetic parameters for M T X plasma data in rabbits following an intra-articular injection of either M T X solution or M T X loaded PLLA2k microspheres (33-110 pm). The dose injected was either 1.5mg or lOmg.  The parameters are  determined using WinNonlin computer program.  MTX  Treatment  tl/2(abs) (h)"  tin (h)  dose (mg)  AUC o-oo  MRT(h)  (fjg/mL.h)  1.5  M T X solution  0.1310.08  1.57+0.90  0.9810.65*  2.2411.47  MTX  0.0410.01  1.0310.33  0.1410.06  1.7010.36  M T X solution  0.0710.05  0.9410.18  4.4012.04*  1.2810.30  MTX  0.0610.03  0.9210.18  0.6510.23  1.3410.23  microspheres 10  microspheres * Statistical difference between two treatments by stduent's r-test (p<0.05). half life ti/2( bs)was estimated by one-compartmental analysis. a  171  a  Absorption  3.3.5  Histological analysis  In an attempt to determine the sites of deposition of the microspheres within the joint, the synovial joint of one of the animals from each treatment group was processed for histological analysis. No synovial proliferation was observed in the joints either injected with either M T X solution or M T X microspheres at either time points (Figure 3.11). For rabbits injected with microspheres, circular empty spaces marked by dark color granules were observed close to the microvessels in the adipose layer of the synovium of the rabbit sacrificed at 6 h after injection (Figure 3.11 C and D).  The  circular spaces were in the size range of 50 pm and appeared to be surrounded by red blood cells in the synovial layer (Figure 3.11 C and D). These circular spaces were not observed in the synovium of rabbits sacrificed 24 h after injection.  172  A)  B)  Figure 3.11 Histological analysis of synovial tissues following intra-articular injection of 56 mg of 18% M T X loaded PLLA2k microspheres (33-110 um) and sacrificed at 6 h following injection. (A) synovial tissue with dark color granules visible in the rectangular area (40x magnification), (B) higher magnification of the rectangular area (400x magnification).  173  Figure 3.12  Histological analysis of synovial tissues following intra-articular injection  of A) 10 mg M T X solution in 400 uL PBS and B) 56 mg of 18% M T X loaded PLLA2k microspheres ( 33-110 urn) and sacrificed at 24 h following injection.  (Magnification  100 x). Both graphs showed no evidence of synovial proliferation as indicated by the arrows.  174  3.3.6  A S T analysis  M T X has been reported to induce acute hepatotoxicty.  In a clinical setting, serum  transaminases, aspartamine transferase (AST) and alanine transferase (ALT) are often monitored (Kirchain and Gill, 1996). The high concentration of these two enzymes and their ready release from the hepatocyte cytoplasm makes them good indicators of necrotic lesions within the liver.  The AST activities in rabbit plasma at different times  following intra-articular injection of lOmg M T X solution or M T X loaded microspheres were determined and the results are shown in Figure 3.13. There was no difference in the A S T activity profiles for rabbits injected with either M T X solution or M T X loaded microspheres.  The A S T in the rabbit plasma increased slightly following the injection  of M T X formulations and peaked at 6 h following the injection. The activity declined 24 h following injection. Although the AST activity slightly increased following the injection of M T X , the activities were within the normal range of A S T (below 28 SF units/mL) (McLauglin and Fish, 1994) indicating that the injection of M T X did not induce acute liver toxicity.  175  Time (h)  Figure 3.13  The A S T activities in rabbit plasma after a single intra-articular injection  of either 10 mg M T X solution in 400 uL PBS or 56mg of 18% M T X loaded PLLA2k microspheres (33-110 pm) in 400 p L PBS. Values are mean ± one standard deviation, (n = 8 for 0-6 h period, n = 4 for 6-24 h period). The dose of M T X injected was 10 mg. (•) M T X solution, (•) M T X microspheres.  176  3.4  DISCUSSION  In both low dose and high dose studies, M T X appeared in the plasma of rabbits injected intra-articularly with either M T X solution or M T X microspheres about 5 min following injection, and peak concentrations occurred at 15 min, indicating that M T X was rapidly transferred from the synovial cavity into the systemic circulation (Figure 3.4 and Figure 3.6).  This rapid transfer of drug was also confirmed by the short absorption  half-life (0.04 to 0.13 h) observed in both groups at both doses (Table 3.6).  A similar  profile to the one obtained after injection of the M T X solution was observed by Foong and Green(1988) who administered a single intra-articular injection of [ H] M T X solution into normal or inflamed joints of rabbits. The maximum concentration was observed at 1 h following the intra-articular of lmg [ H] M T X solution (Foong and Green, 3  1988). The rapid appearance of M T X in the plasma of rabbit injected with M T X microspheres was likely due to the burst phase of M T X release from microspheres observed in previous in vitro drug release studies (Figure 2.4).  In both the high and low  dose studies, plasma levels of M T X attained following intra-articular injection of M T X microspheres were more than 5 fold lower than levels produced by M T X solution (Figure 3.4 and Figure 3.6).  After about 2 h post intra-articular injection of microspheres,  177  plasma M T X concentrations were very low and below detection level 3-6 h following microspheres injection.  The in vitro M T X release profiles of 10% and 20% (w/w) M T X  loaded microspheres have demonstrated a rapid burst phase of release of M T X in the first 4 h followed by a slower release period (Figure 2.4).  We speculate that the release of  M T X from microspheres in the joint cavity slowed down 2 h following intra-articular injection due to the ending of the burst release phase and possible aggregation and adherence of microspheres to the synovial tissues resulting in low synovial fluid and plasma concentrations (Nishide et al., 1999). When comparing the pharmacokinetic parameters for plasma M T X data from both studies, the AUCo-oo of the groups injected with M T X solution was about 7 fold higher than the AUCo-oo of the groups injected with M T X microspheres (Table 3.6) indicating that more drug entered the systemic circulation from the groups injected with the M T X solution.  The Cmax (Figure 3.7) and AUCo-oo values (Table 3.6) for the 10 mg M T X  solution plasma concentration-time curve were between 5-7 fold greater than the Cmax (Figure 3.5) and AUCo-oo (Table 3.6) values for the 1.5 mg M T X solution plasma concentration-time curve.  This suggested that plasma levels following intra-articular  injection were dose dependent and that the transfer of M T X solution from the synovial cavity to the systemic circulation did not reach saturation in the high dose study.  178  The  mean residence times (MRT) and the elimination half-lives tm for both treatment groups at two doses were not significantly different (Table 3.6).  Since M R T after an  extravascular administration is a function of both drug absorption and elimination, the fact that the M R T values were similar in both treatment groups, confirms that the rates of absorption were similar. The elimination half-life of M T X following systemic administration in rabbits, reported in the literature, varies from 2 to 47 h depending on the dose, the duration of the study and the detection limit of the assay(Sasaki et al, 1983; Chen and Chiou, 1983; Iven et al, 1985). The half-lives of M T X have been reported as 0.7 h (oti/2) and 47.7 h (p\i/2) t  following a 6-h 50mg/kg intravenous infusion (Sasaki et al., 1983).  Chen and Chiou  (1983) showed that following an intravenous bolus of a dose of 45 mg M T X in rabbits, M T X plasma levels decreased polyexponentially with an elimination half-life of 10 h. Iven et al (1985) found that following a short term infusion of 1.33 mg/kg M T X for 10 min in rabbits, the plasma concentrations decreased in a triexponential manner with a i/2, t  Pti/2 and yti/2 of 0.136 h, 0.632 h, and 2.402 h, respectively. In this study, plasma was sampled up to 48 h post injection and it is likely that the terminal phase had not been attained.  179  In the high dose study, 7-OH-MTX appeared in the plasma within 5 min and peaked between 1 to 2 h following intra-articular injection in both treatment groups (Figure 3.8A and Figure 3.8B) indicating the injected M T X was quickly converted to its metabolite 7-OH-MTX.  In both treatment groups, there were two rabbits (#5 and #15 for M T X  solution group) and (# 4 and # 14 for microspheres group) that possessed markedly different 7-OH-MTX plasma profiles than the rest of the rabbits (Figures 3.7 and 3.8). These differences in the plasma profiles of 7-OH-MTX could be due to individual variation in hepatic MTX-7-hydroxylase activities. A 48 fold range in MTX-7-hydroxylase activities in six human liver cytosol preparations have been shown by Kitamura et al (1999). In the low dose study, during the period of 8-24 hours urine collection, 8% of the total drug was excreted in the urine following injection of M T X solution compared to 2% of the drug excreted from rabbits injected with M T X loaded microspheres (Figure 3.5). This result was consistent with the plasma data indicating that M T X from the solution formulation rapidly entered the systemic circulation and was eventually excreted in the urine, while the uptake of M T X from the joint cavity into the blood was slowed down in the microspheres formulation.  In the period of 24-48 h, an additional 3% of M T X was  excreted from rabbits injected with M T X loaded microspheres (Figure 3.5) suggesting  180  that M T X was still being released at a very slow rate from microspheres, despite the fact that the concentration of M T X in plasma was below detection level 4 hours after intra-articular injection.  The in vitro release profile of 10% (w/w) loaded M T X  microspheres have shown that following the burst phase of M T X release, the loaded M T X was released at an approximately zero order rate of 0.2% per day over 2 weeks (Figure 2.4). Urine data showed that 2.5 times more M T X was excreted in the urine from rabbits injected with M T X solution than rabbits injected with M T X loaded microspheres 48 h after intra-articular injection (Figure 3.5).  For the low dose study, the  total amount of M T X excreted in the urine at the end of the study accounted for 10% of total dose injected for rabbits injected with M T X solution and 5% for rabbits injected with M T X loaded microspheres (Figure 3.5). Due to the fact that we failed to collect urine samples in the first two hours following intra-articular injection, the urine data do not completely represent the total amount of M T X excreted in the urine. In the low dose study, the synovial joints were not flushed with saline and the whole right joints of the rabbits were removed for M T X analysis. Approximately 0.60 pg of M T X was detected in the joint tissues of two of the rabbits injected with M T X microspheres while no M T X was detected in the joint of rabbits injected with M T X solution.  The fact that we were not able to detect significant amounts of M T X in the  181  joints of rabbits injected with M T X microspheres was likely due to the large sample size of joint tissues (8-13 g) taken for analysis. The whole joint was homogenized, extracted and samples analyzed by H P L C .  We speculate that the concentrations of M T X in the  samples were below the quantitation limit of the assay. In the high dose study, the rabbits were under light inhalation anesthesia and a catheter was inserted into the urethra of the rabbits to ensure the complete collection of urine samples. At 0-3 h following intra-articular injection, the amounts of M T X and 7-OH-MTX excreted in the urine of rabbits injected with solution were approximately 6 fold higher than in the urine of rabbits injected with M T X loaded microspheres (Figure 3.9A and B).  Within 24 h following injection, 4-5 fold more M T X and 7-OH-MTX was  excreted in the urine from rabbits injected with M T X solution than rabbits injected with M T X loaded microspheres (Figure 3.9A and Figure 3.9B). Both high and low dose studies showed that following intra-articular injection, M T X from M T X solution rapidly entered the systemic circulation and was excreted in the urine while M T X microspheres slowed down the uptake of M T X across synovium into blood circulation by releasing M T X in the joint in a slow and controlled manner. In the high dose study, the synovial joints were flushed with heparinized saline in an attempt to recover the microspheres and M T X in the synovial fluid.  182  At 6 h post  injection, the amount of M T X in aspirated fluid of rabbits injected with microspheres was significantly higher than that of the rabbits injected with M T X solution, suggesting that M T X microspheres were able to retain the drug in the joint (Figure 3.1 OA).  At 24 h  following intra-articular injection, no drug was detected in the aspirated fluid of rabbits administered M T X solution, while small amounts of M T X were detected in the aspirated fluid of the rabbits injected with microspheres (Figure 3.1 OA).  We speculate that by 24  h following injection, some M T X microspheres may have became located in the joint in such a way that they were inaccessible to the joint flushing procedure.  The significant  amounts of M T X detected in synovial tissues at 24 h (Fig 3.10 B) may indicate that microspheres migrated to popliteal folds of the synovial cavity where they were not flushed out with saline but were extracted in the synovial tissue samples.  In studies  where rabbit joints were injected with microspheres (10 mg M T X dose) and then immediately flushed and fluids aspirated to recover the microspheres, only about 10-13% of the total injected dose was recovered (Table 3.5). This study illustrated the difficulty in retrieving the microspheres and synovial fluid from the joint. In the high dose study, for rabbits injected with M T X solution, 72% of the total dose was excreted in the urine (28% as M T X and 44% as 7-OH-MTX) and approximately 0.1% of the drug was detected in the synovial tissues and kidneys (Figure 3.1 OA and  183  Figure 3.1 OB) 24 h following intra-articular injection.  M T X was not detected in other  tissues removed for analysis. For rabbits injected with M T X loaded microspheres, 12% of total dose was excreted in the urine (5% as M T X and 7% as 7-OH-MTX) 24 h following intra-articular injection of M T X loaded microspheres and approximately 2% of the total M T X was recovered from synovial tissues and synovial fluid (Figure 3.1 OA and Figure 3.1 OB). No M T X was detected in other tissues.  It is possible that the rest of the  injected M T X that was not detected in the rabbits may have been metabolized into M T X polyglutamates intracellularly (Iven et ah, 1985) which could not be analysed by the HPLC assay, and also distributed in the tissues and plasma at concentrations below the detection limit of the assay.  It has been shown that M T X was rapidly converted to M T X  polyglutamates and accumulated in the brain, liver, kidneys, and testes of rats as early as 3 h following an intra-peritoneal injection of lOmg/kg of M T X (Krakower and Kamen, 1983). The accumulation of M T X polyglutamates has also been demonstrated in erythrocytes, T-lymphocytes, fibroblasts, and hepatic tissues of patients with rheumatoid arthritis receiving long term methotrexate therapy (Bannwarth et al., 1996; Grim et al., 2003). Renal excretion is the major route of elimination for M T X and its metabolite 7-hydroxy-methotrexate (7-OH-MTX) in rabbits (Chen and Chiou, 1983). Chen and  184  Chiou (1983) reported that 67% of drug was excreted as M T X and 33% as 7-OH-MTX in the urine of rabbits 48 h after an intravenous bolus injection of 45mg of M T X . Administration of a 6 h 50mg/kg infusion in rabbits showed that 57% of the dose (45% as M T X and 10% as 7-OH-MTX) was excreted in the urine in 20 h with 10%, 1% and 1.8% of total dose in the tissues, G l tract, and bile, respectively (Sasaki et al., 1983). For the tissue samples, liver, kidney, and lungs possessed the highest concentrations (Sasaki et al., 1983). Foong and Green (1988) showed that following an intra-articular injection of lmg [ H]MTX solution in rabbits, 79% of the total radioactivity was recovered in the 3  urine, while 0.41% and 0.14% were recovered from liver and kidney, respectively, 24 h following injection. The histological analysis showed that at 6 h post intra-articular injection of M T X loaded microspheres, there were empty spaces surrounded by dark color granules embedded in theadipose layer of the synovium (Figure 3.11A and B).  Considering the  similarity in sizes and shapes, these spaces were likely caused by tissue deposition of the microspheres that were subsequently removed by the processing conditions during the fixation procedures. synovial cells.  There was no evidence of microsphere phagocytosis by the  Horisawa et al (2002a) administered intra-articular injections of P L G A  microspheres with a mean diameter of 26.5 um in rat joints and demonstrated the  185  presence of a granulation tissue surrounded by multinucleated giant cells in the synovium.  A possible explanation for the deposition of microspheres in the adipose  layer observed in this work was the creation of turbulence brought about by intra-articular injection.  Since the adipose layer is only 10-15 urn away from the  synovial surface, the relatively high flow rate accompanying the intra-articular injection could have formed a channel in the tissue and carried the microspheres to the adipose layer. The accumulated red blood cells surrounding the spaces created by microspheres likely leaked from nearby microvessels due to the trauma.  Despite the possible trauma  caused by the initial intra-articular injection, no synovial proliferation was observed indicating the deposition of microspheres in the joint cavity did not induce acute injury (Figure 3.12 A and B).  No evidence of microspheres deposition was found for the rabbit  sacrificed at 24 h following injection. We speculate that since only one joint was taken from 4 treated animals and only 8 sagittal slides were cut across the synovial joint, that it is possible that microspheres migrated to the popliteal fold as suggested by Nishide et al (1999) and the section of the fold that contained the microspheres was missed. In summary, M T X loaded PLLA2k microspheres were retained in the joint space and synovium and released M T X in a slow, controlled manner.  M T X in solution was  rapidly absorbed across the synovium into the plasma and systemic uptake of M T X was  186  considerably higher than for M T X microspheres.  M T X levels in synovial tissues, 24 h  after intra-articular injection were markedly higher for the microspheres formulation compared to the solution formulation.  187  Chapter 4 EFFICACY OF INTRA-ARTICULAR MTX LOADED MICROSPHERES 4.1  INTRODUCTION  In the previous chapter, it was shown that intra-articular injections of M T X loaded microspheres into healthy rabbits resulted in low systemic levels of drug compared to M T X solutions. Urinary excretion data for M T X and 7-OH-MTX and synovial tissue levels of M T X suggested that M T X was released by the microspheres in the joint at low levels over at least 48 h. Rabbits are generally considered to be a good animal model for studies of rheumatoid arthritis and for the evaluation of efficacy of anti-rheumatic drugs (Foong and Green, 1988). Dumonde and Glynn (1962) reported the induction of chronic synovitis in sensitized rabbits by the intra-articular injection of fibrin.  Clinically and  histologically acute, and later chronic, monoarthritis was produced, lasting for months and resembling rheumatoid arthritis synovitis. Lining cell hyperplasia, perivascular diffuse infiltration with lymphocytes, plasma cells, lymphoid follicles, pannus and cartilage erosions were the prominent features indistinguishable from those found in rheumatoid arthritic synovitis.  Since then, antigen-induced arthritis has been one of the  most intensively studied animal models of inflammatory arthritis (Magilavy, 1990).  188  Various antigens including bovine serum albumin, ovalbumin, and horseradish peroxidase are all capable of inducing infiltration of neutrophils, plasma cells, macrophages and lymphocytes (Magilavy, 1990). By injecting antigen locally into the joints of presensitized animals, monoarthritis induction is possible, as compared to polyarthritis in other animal models (Brahn, 1991). This model is particularly useful for the evaluation of intra-articular treatments in which the treatment is localized in one or two joints. Evidence has shown that antigen-induced arthritis in rabbits and rats is responsive to both intra-articular and systemic M T X treatment. Foong and Green (1993) induced arthritis in ovalbumin-presensitized rabbits with 5mg of ovalbumin in both knee joints.  M T X solution or a M T X liposomal formulation was injected into one of the knee  joints of rabbits, either immediately after antigen challenge (day 0), day 7, day 21 or day 35 and the progression of the disease was monitored for 56 days.  The results showed  that the rabbits treated with lmg M T X solution at day 0 and rabbits treated with the M T X liposomal formulation at day 0 and day 7 had less joint swelling, lower skin temperature, decreased synovial fluid production and synovium proliferation compared to the contra-lateral controls.  The intra-articular injection of M T X solution or the M T X  liposomal formulation administered several days after antigen challenge had no  189  significant effect on the established arthritis (Foong and Green, 1993). Williams et al (1996) evaluated the efficacy of intra-articular delivery of 200 pg M T X in an antigen-induced arthritis model in rats. M T X solution or a M T X liposomal formulation were given 7 days following the induction of arthritis and the progression was monitored for 21 days.  No significant differences in the development of arthritis were observed in  the M T X solution treated rats compared with saline treated control rats, while a rapid reduction in knee swelling within 24 h and a progressive reduction in joint swelling over the next 20 days was observed in the animals treated with the M T X liposomal formulation.  The overall inflammatory scores in M T X liposomal formulation treated  rats were significantly lower than those in saline-treated controls (Williams et al., 1996). Williams et al (2001) also evaluated the effects of a M T X liposomal formulation on cytokine mRNA expression following intra-articular injection; On the day of the arthritis challenge, rats were treated with a single intra-articular injection of M T X liposomal formulation, or M T X solution. On day 3 and day 7 after disease induction, animals were sacrificed. The results showed that the M T X liposomal formulation significantly reduced knee swelling by day 1 and inhibited the histological progression compared to M T X solution. The local proinflammatory cytokines IL-1 (3 and IL-6 mRNA expression in synovial tissue extracts were also reduced (Williams et al., 1995;  190  Wiilliams et al, 2001). Novaes et al (1996) showed that a low dose intramuscular treatment of M T X at 0.25mg/kg to rabbits 1 week before the induction of arthritis in the knee joint, significantly reduced the intensity of leukocyte influx, protein leakage, synovial membrane cell infiltrate, as well as the production of IL-1 p cytokines. The antigen-induced arthritis model has also been used to evaluate other intra-articular anti-rheumatic and anti-inflammatory drug formulations.  Horisawa et al  (2002b) have shown that betamethasone encapsulated nanospheres injected intra-articularly into the joints of antigen- induced arthritis rabbits produced a reduction in joint swelling and cell infiltration compared to betamethasone solution and saline controls. Liggins et al (2004) showed that paclitaxel loaded microspheres injected intra-articularly in an antigen-induced arthritis rabbit model, produced a greater reduction in joint swelling, cellular infiltration, and cartilage damage compared to control microspheres. In this work, the in vivo efficacy of intra-articular M T X microspheres was evaluated in the antigen-induced arthritis rabbit model, using ovalbumin as the antigen. The objectives of this work were: 1. To evaluate the efficacy of intra-articular M T X microspheres or M T X solution in rabbits with antigen-induced arthritis in hind knee joints.  191  2. To determine the pharmacokinetics of M T X in the antigen-induced arthritis rabbit model following intra-articular injection of M T X loaded microspheres or M T X solution.  192  4.2 4.2.1  EXPERIMENTAL Materials  Methotrexate (MTX) (MW: 454.4g/mole) was purchased from Hande Tech Development Co. (U.S.A).  Aminopterin was purchased from Sigma Chemical Co. Poly  (vinyl alcohol) (PVA) (98% hydrated, M W : 13,000-23,000) was obtained from Aldrich Chemical Company Inc. 7-hydroxy-methotrexate (7-OH-MTX) was purchased from Schircks Laboratories (Jona, Switzerland). Poly(L-lactic acid) (MW: 1600-2400 g/mol, intrinsic viscosity: 0.1-0.2, polydispersity: 2-3) was obtained from Polysciences (Warrington, PA). (HPLC) grade.  A l l solvents used were High Performance Liquid Chromatography  Phosphate buffered,saline (PBS, pH 7.4 ) was prepared by dissolving  0.32g sodium dihydrogen orthophosphate, 2.15 g sodium phosphate, 8.22 g NaCl in one liter of distilled water.  Phosphate buffer (pH 6.5) was prepared by dissolving 0.936 g  sodium dihydrogen orthophosphate, 0.644 g sodium phosphate in one liter of distilled water and filtered through 0.2 mm filter paper.  18% M T X loaded P L L A microspheres  (33-110 pm) were from the same batch of microspheres used for the high dose pharmacokinetic study previously reported in Section 3.2.2.  193  4.2.2  Animals and housing  Twelve female New Zealand White rabbits (weight 4-5 kg) were obtained from Riemens farm (Calgary, Canada).  The experiments were conducted at the Animal  Resource Center of the University of Calgary and the animals were housed in metal cages. The room was maintained at a temperature of approximately 20°C with 30-70% relative humidity and a light/dark cycle of 12 hours/12 hours.  Rabbit chow and tap  water were provided ad libitum to the animals for the duration of the study.  4.2.3  Animal care and use committee approval  This study fell under the scope of protocol No. M03147 approved by the Committee on Animal Care of the University of Calgary.  The experiments were conducted in  accordance with the principles contained in Care of Experimental Animals - A Guide for Canada, published by the Canadian Council on Animal Care.  4.2.4  Induction of rheumatoid arthritis  Arthritis was induced in the knee joints of rabbits using a procedure similar to that described by Foong and Green (1988).  Twelve female New Zealand White rabbits were  immunized by injecting subcutaneously a total of 1 mL of 10 mg/mL ovalbumin (Sigma, ON, Canada) in sterile PBS emulsified in Complete Freund's Adjuvant (Difco, Fisher Scientific Inc.).  Fourteen days later, a booster of 1 mL of 10 mg/mL ovalbumin in 194  sterile PBS and Incomplete Freund's adjuvant (Difco, Fisher Scientific Inc.) were given subcutaneously.  Twenty-eight days after the first immunization, arthritis was induced  with a 0.5 mL intra-articular injection of 10 mg/mL ovalbumin in sterile PBS into the suprapatellar pouch of both left and right hind knee joints using the medial approach while animals were anaesthetized with 1.5%- 4% inhalation halothane.  4.2.5  Intra-articular injection and blood collection from the rabbits  Twenty-four hours following the induction of arthritis, the animals received treatments with M T X .  The animals were under 1.5%- 4% inhalation halothane  anesthesia throughout the process. The right knee joints of treated rabbits received an injection of either 10 mg M T X solution in 400 uL sterile PBS (pH 7.4) (n=5) or 56 mg of 18% M T X loaded P L L A microspheres (33-110 pm) in 400 uL sterile PBS (n=5). The M T X stock solution was prepared by dissolving 100 mg M T X in 1 mL Sodium Hydroxide and dilutes it with PBS to 4 mL. 400 pL sterile PBS.  The left knee joints received an injection of  Control animals received an injection of 400 pL sterile PBS in both  knees (n=2). The detailed assignments of intra-articular treatments to individual rabbits are given in Table 4.1. Due to the longer duration (up to 14 days) of the experiment, the use of a jugular catheter was not approved by the Animal Care Committee in the University of Calgary.  195  Depending upon the ease with which blood could be obtained from individual animals, blood samples were taken from either the ear vein or the ear artery of both ears of the rabbits using a needled catheter. Immediately before (0 h) and following the intra-articular injection, serial blood samples (0.5 mL) were obtained at 0.5, 1, 2, 4, 6, 8 h to study the pharmacokinetics of M T X following intra-articular injection.  Table 4.1  Intra-articular treatments to individual rabbits one day following the  induction of arthritis by ovalbumin in both knee joints.  PBS: 400 uL sterile PBS, M T X  microspheres: 56 mg of 18% M T X loaded PLLA2k microspheres in 400 u.L sterile PBS, M T X solution: 10 mg M T X solution in 400 pL sterile PBS. The dose of M T X injected was 10 mg. Intra-articular Treatments  Rabbit #  Left knee  Right knee 1  PBS  PBS  2  M T X microspheres  PBS  3  M T X solution  PBS  4  M T X microspheres  PBS  5  M T X solution  PBS  6  M T X microspheres  PBS  7  M T X solution  PBS  8  PBS  PBS  9  M T X solution  PBS  10  M T X microspheres  PBS  11  M T X solution  PBS  12  M T X microspheres  PBS  196  4.2.6  Analysis of M T X and 7 - O H - M T X in rabbit plasma  M T X and the metabolite 7-OH-MTX in the plasma samples were extracted according to the method previously reported in Section 3.2.3.1.  Briefly, 1 mL  acetonitrile was added to the plasma sample (0.5 mL) vortex mixed and centrifuged at 16000 x g for 5 minutes.  The supernatant was transferred to a 12mL glass tube and 2mL  of methylene chloride were added to extract the aqueous phase. The aqueous phase (200 pL) was dried under a stream of nitrogen gas at 45°C.  The dried sample was  reconstituted with 200 uL phosphate buffer pH 6.5.  4.2.7  HPLC analysis of M T X and 7 - O H - M T X  M T X and 7-OH-MTX in plasma were assayed by H P L C using the method reported previously in Section 3.2.3.2.  The H P L C system consisted of a Shimadzu LC-10AD  pump (Shimadzu Corporation, Japan) and a Shimadzu SIL-9A automatic injector.  A  Waters 470 scanning fluorescence detector was used (Ex: 370 nm, Em: 417 nm). A photochemical reactor unit with a 254 nm U V lamp and a 10m x 0.25mm reaction coil (Aura Industries Inc, N Y , USA) was used post-column (before the detector). The analytical column was a Novo-Pak C18 column with dimensions of 3.9 x 150 mm. The mobile phase consisted of 0.01M phosphate buffer pH 6.5, with 3.2% acetonitrile and 0.2% of 30%) hydrogen peroxide (Sigma Chemical Co.) at a flow rate of l m L /min under 197  ambient temperature.  The injection volume was 50 pL.  A folic acid antagonist  aminopterin at the concentration of 0.2 ug/mL was used as the internal standard and calibration curves were constructed by plotting the ratio of peak areas of MTX/internal standard versus concentrations of M T X .  4.2.8  Pharmacokinetic calculations  The plasma curves were resolved with the nonlinear regression computer program WinNONLTN (Scientific Consulting Inc., Standard Edition, Version 1.1) using the non-compartmental and the one-compartmental model as described in Section 3.2.6. The elimination half-life (ti/2), the area under the curve (AUCo-oo) and the mean residence time (MRT) were determined by using Equations 3.1, 3.2 and 3.4, respectively. The absorption half-life  4.2.9  ti/2( bs) a  was determined by Equations 3.5 and 3.6.  Statistical analysis Student's t test (p<0.05) was used to compare pharmacokinetic parameters between  two treatments. 4.2.10  Monitoring of arthritis development  The inflammatory response was monitored by measuring changes in the joint diameter by a digital caliper (VWR, Canada). Measurements were taken when the hind leg was at an extended angle and flexed at a 90° angle. The differences in these two  198  measurements were in the range of 0.1 mm and therefore, subsequent measurements were only taken at the 90° angle.  The development of arthritis was monitored for 14 days by  measuring the diameter of the joint daily for the first 7 days and then every other day. Fourteen days post intra-articular treatment; all the rabbits were humanely sacrificed and the whole knee joints were taken for standard histological analysis.  4.2.11  Histological processing of knee joints  Histological processing was carried by technical staff at the Faculty of Medicine, University of Calgary.  Rabbit knee joints were transected 3-4 cm above and below the  knee joint. Muscle was stripped and the joint was fixed in 10 % Neutral Buffered Formalin (Fisher Scientific Inc.) for 10 days at room temperature.  Joints were  transferred to Cal x II decalcifying solution, which consisted of 10% formic acid solution in 4% formaldehyde (Fisher Scientific) and kept at room temperature. The solution was changed daily for 14 days at which time the joint was bisected in the sagittal plane. Joints were left for a further 14 days with changes of decalcifying solution every 2 days. The end point was decided when bones could be cut smoothly with a blade without any grittiness. 4.1.  The gross appearance of a decalcified rabbit knee joint is shown in Figure  The tissues were then washed thoroughly under cool running tap water for 2 hours,  before being transferred to an automatic paraffin processor (Leica TP 1020). Routine  199  processing for histological analysis including dehydration in ethanol, clearing in xylene and infiltration in Paraplast Plus Wax (Fisher Scientific Inc.) was accomplished overnight.  The tissue was then transferred to a vacuum oven at 60°C for 2 hrs, to ensure  complete infiltration of paraffin into the tissue.  Bones were embedded in paraffin  moulds and stored at room temperature until sectioned. Serial sections were cut 12 pm in thickness using a Leica R M 2165 microtome. Sections were adhered to Superfrost Plus slides (Fisher Scientific) and baked at 40 °C for 5 days.  Alternate slides were then stained with haematoxylin, fast green and safranin O  (Fisher Scientific Inc.). Sections was cleared in xylene, cover slipped with Polymount mounting media (Fisher Scientific Inc.) and allowed to dry at room temperature for several days.  200  A)  Figure 4.1  B)  The gross appearance of a decalcified rabbit knee joint. A) Rabbit knee  joint transected 3-4 cm above and below the knee joint. B) The joint was bisected in the sagittal plane. C) The gross appearance of the knee joint following bisection.  201  4.2.12  Grading histological slides  Histological grading was carried out by experienced histologists from the Faculty of Medicine, University of Calgary. A l l sections were coded prior to assessment and graded in a blinded fashion to eliminate observer bias.  Since there are still no standard  criteria for evaluating histopathology of rheumatoid arthritis, a list of criteria was compiled and modified from those previously reported (Kapila et al., 1995; Mould et al., 2003) and is provided in Table 4.2.  First, all sections were viewed under an optical  microscope (Leitz D M R B ) and examined for typical defects of rheumatoid arthritis and a score was determined for each defect based on the criteria. Then every 10 section was th  examined and graded again to compare scores.  Finally, 3 slides from both medial and  lateral compartments of the joint were selected at similar locations on each joint and again graded by the same person. observer.  The scores were validated by a second independent  The total score was determined by adding the scores for the 6 histological  criteria.  202  Table 4.2  Scoring criteria for histological analysis of rabbit joints induced with  arthritis by ovalbumin and intra-articularly treated with M T X . The list of criteria was compiled and modified based on those previous reported (Kapila et al., 1995; Mould et al, 2003). Histopathologicalfeatures  Score  Synovial lining hyperplasia 1-3 layers  0  4-6 layers  1  > 7 layers  2 Villous hyperplasia  Not present  0  Few scattered and short villi  1  Marked finger-like villi  2  Marked diffuse villi  3 Inflammatory cell infiltrate  Normal cellularity  0  Increased cellularity  1  Patchy discontinuous cellular infiltration  2  Diffuse continuous cellular infiltration  3  Inflammatory exudates No exudates present  0  Mild exudates present intra-articularly  1  Pannus formation No pannus formation  0  Pannus with superficial cartilage destruction  1  Pannus with destruction to depth of mid zone  2  Pannus with destruction to depth of tide mark  3  Cartilage destruction Cartilage intact  0  Minimal abrasion with no obvious loss of chondrocytes or collagen disruption  1  Destruction with superficial loss of chondrocytes and/or collagen disruption  2  Moderate loss of chondrocytes and collagen disruption to mid zone  3  Destruction with severe loss of chondrocytes to tide mark  4  203  4.3 R E S U L T S 4.3.1  M T X and 7 - O H - M T X plasma concentrations  A n indwelling catheter was not approved for use in these long-term animal studies and blood samples were taken at each time point following intra-articular injection using a needled catheter from either the ear vein or artery. It was difficult to take blood samples via these routes and an average of 10 min was required to remove sufficient blood for a sample. Thus, samples could not be taken at either 5 min or 15 min, as for previous pharmacokinetic studies. The M T X plasma profiles following intra-articular injection of either 10 mg M T X solution or 56 mg of 18% loaded microspheres are shown in Figure 4.2.  Significant  variability between different animals was observed in both treatment groups. Hence, individual M T X plasma profiles for the group treated with M T X solution and the group treated with M T X microspheres are given in Figure 4.3 and Figure 4.4, respectively. In the group treated with M T X solution, rabbits #7, #9 and #11 showed maximal M T X plasma concentrations within 2 h following intra-articular injection of the microspheres (Figure 4.3). M T X was still detectable 24 h following injection.  Two rabbits #3 and  #5, showed peak M T X concentrations at between 30 min to 1 h following intra-articular injection of M T X and peak plasma concentrations were about 10 fold higher than rabbits  204  #7, #9, and #11.  Rabbit #5 showed a prolonged phase of elimination compared to all  other rabbits. For rabbits injected with M T X microspheres, the inter-individual variability in M T X plasma profiles was not as large as in the group treated with M T X solution. The peak M T X concentrations in the group treated with M T X microspheres ranged from about 0.1 to 0.3 pg/mL and gradually declined to undetectable levels 8 h following the injection (Figure 4.4). The 7-OH-MTX plasma profiles of rabbits injected with either M T X solution or M T X loaded microspheres are shown in Figure 4.5 and 4.6.  For rabbits injected with  M T X solution, the concentration of 7-OH-MTX peaked between 2 to 3 h and peak concentrations were between 0.5 and 1.3 pg/mL.  Except in one rabbit (#5), the  concentrations of 7-OH-MTX were higher than M T X in plasma from 2 to 8 h following injection (Figure 4.5). For rabbits injected with M T X microspheres, the concentrations of 7-OH-MTX peaked between 0.5 to 2 h following injection, and peak concentrations of 7-OH-MTX were between 0.1 and 0.3 pg/mL (Figure 4.6). The pharmacokinetic analysis of the M T X plasma profiles following the intra-articular injections of M T X solution or M T X microspheres into inflamed joints are shown in Table 4.3. Data from normal joints previously reported in Table 3.6 are shown for ease of comparison.  For rabbits with antigen induced arthritis, the area under the  205  curve (AUC o-«0 of the M T X plasma profile of rabbits injected with M T X solution was significantly higher than the M T X plasma profile for rabbits injected with M T X microspheres.  The elimination half-lives (ti/2) were 6.7 h and 2.5 h for rabbits injected  with M T X solution and M T X microspheres, respectively.  The differences observed in  elimination half-lives (ti/2) and mean residence time (MRT) between two treatment groups were not statistically significant. Comparing diseased and normal animals, the M T X mean absorption half-lives (t\a  (abs))  and elimination half-lives (ti/2) for diseased  animals in both treatment groups were longer than those of the healthy animals. However, statistical analysis of the data did not show any significant difference due to the large standard deviations.  206  2.5  2 "5b  HL  1.5  a "E. X H  0.5  10  15  20  25  30  Time (h)  Figure 4.2  M T X concentrations in plasma of antigen (ovalbumin) induced arthritis  rabbits following a single intra-articular injection of either 10 mg M T X solution in 400 uL PBS or 56 mg of 18% PLLA2k M T X loaded microspheres (33-110 urn) in 400 uL PBS into the right hind knee joint.  Left hind knee joints were injected with 400 pL PBS.  Values are mean ± one standard deviation, (n = 5). The dose of M T X injected was lOmg. (•) M T X solution, (•) M T X microspheres.  207  0  5  10  15  20  25  30  Time (h)  Figure 4.3  M T X plasma concentrations of individual antigen (ovalbumin) induced  arthritis rabbits following a single intra-articular injection of 10 mg M T X solution in 400 uL PBS into the right hind knee joint. Left hind knee joints were injected with 400 uL PBS.  (•) rabbit #3, (•) rabbit #5, ( A ) rabbit #7, (x) rabbit #9, (•) rabbit #11.  208  0.35 r  Time (h)  Figure 4.4  M T X plasma concentrations of individual antigen (ovalbumin) induced  arthritis rabbits following a single intra-articular injection of 56 mg of 18% M T X loaded PLLA2k microspheres (33-110 pm) in 400 pL PBS into the right hind knee joint. hind knee joints were injected with 400 uL PBS. MTX.  The dose of M T X injected was 10 mg  (•) rabbit #2, (•) rabbit #4, ( A ) rabbit #6, (x) rabbit #10, (•) rabbit #12.  209  Left  1.6  r  1.4 -  0  5  10  15  20  25  30  • Time (h)  Figure 4.5  7-OH-MTX plasma concentrations of individual antigen (ovalbumin)  induced arthritis rabbits following a single intra-articular injection of 10 mg M T X solution in 400 pL PBS into the right hind knee joint. injected with 400 uL PBS.  Left hind knee joints were  (•) rabbit #3, (•) rabbit #5, ( • ) rabbit #7, (x) rabbit #9,  (•) rabbit #11.  210  0.35 -,  0  2  4  6  8  10  Time (h)  Figure 4.6  7-OH-MTX plasma concentrations of individual antigen (ovalbumin)  induced arthritis rabbits following a single intra-articular injection of 56 mg of 18% M T X loaded PLLA2k microspheres in 400 uL PBS to the right hind knee joint. knee joints were injected with 400 uL PBS. MTX.  Left hind  The dose of M T X injected was 10 mg  <•) rabbit #2, (•) rabbit #4, (A) rabbit #6, (x) rabbit # 10, (•) rabbit #12.  211  Table 4.3  Pharmacokinetic parameters for M T X plasma profiles of antigen  (ovalbumin) induced arthritis rabbits following an intra-articular injection of either M T X solution or M T X loaded microspheres.  The dose of M T X injected was lOmg. Data  from normal rabbits injected with the same amount of M T X solution and M T X microspheres are shown here for ease of comparison (from Table 3.6).  The parameters  were determined using WinNonlin computer program.  Treatment  tl/2(abs) (hf  ti/ (h) 2  AUC o-co  MRT(h)  (fig/mL.h) Inflamed joints M T X solution  0.59 ±0.46  6.72 ± 4.62  6.99 ± 7.20*  8.81 ±5.96  M T X microspheres  0.31 ±0.23  2.54 ± 1.79  0.47 ± 0 . 1 8  3.92 ±2.68  M T X solution  0.07 ±0.05  0.94 ±0.18  4.40 ± 2.04*  1.28 ±0.30  M T X microspheres  0.06 ± 0.03  0.92 ±0.18  0.65 ± 0.23  1.34 ±0.23  Normal joints  *Statistical difference between two treatments by student's r-test (p<0.05). a  Absorption half life ti/2( bs)was estimated by one-compartmental analysis. a  212  4.3.2  Joint swelling evaluation  The joint diameters of the rabbits were measured using a digital caliper.  Except  on day 0 and day 1, the rabbits were not under anesthesia when the knee joint diameter was measured. The results of knee swelling measurements (knee diameter on each day minus knee diameter on day 0) are provided in Figure 4.6 (A-C).  The joint swelling  consistently increased in the first three days following the injection of M T X solution or M T X microspheres and decreased on day 4. The swelling then peaked again at 7 days following the injection of M T X solution or microspheres and gradually declined to similar level as day 1. There were no significant differences between joints injected with M T X microspheres, M T X solution or PBS.  213  A)  15  r  0  2  4  6  8  10  12  Days following induction of arthritis  214  14  16  Figure 4.7  Knee joint swelling following antigen (ovalbumin) induction of arthritis in  rabbits. One day following the induction of arthritis, the right hind knees of the rabbits were injected with either 10 mg M T X solution in 400 uL PBS or 56 mg of 18% M T X loaded microspheres (33-110 pm) in 400 u.L PBS. injected with 400uL PBS.  The left hind knee joints were  A ) Comparison of knee joint swelling between M T X  solution treated knees and 400uL PBS treated contralateral knees, B) comparison of knee joint swelling between M T X loaded microspheres treated and PBS treated contralateral knees, C) comparison of knee joint swelling between M T X solution treated knees and M T X microspheres treated knees.  Values are mean ± one standard deviation, (n = 5).  The dose of M T X injected was lOmg.  (•) M T X solution, (•) M T X microspheres, (x)  PBS.  215  4.3.3  Histological Analysis  Fourteen days following the intra-articular treatment with M T X solution or M T X microspheres, the rabbits were sacrificed and the whole joints were isolated and processed for histological analysis. Histological slides were scored in five categories: synovial lining hyperplasia, villous hyperplasia, inflammatory cell infiltrate, inflammatory exudates, pannus formation and cartilage destruction.  These histopathological features were clearly  observed in the rabbits induced with arthritis suggesting that arthritis was successfully induced in this animal model (Figure 4.10-Figure 4.16).  Optical micrographs of the  cartilage of a healthy rabbit knee joint obtained from another independent study are shown in Figure 4.8 for comparison. In the healthy joint, the cartilage surfaces covering femur and tibia are smooth with the different zones that are commonly used to describe the cartilage, clearly visible (Figure 4.8B).  Zone I is the superficial zone of the cartilage  that only comprises 1-3 layers of chondrocytes.  Zone II is the transitional zone where  oval or rounded chondrocytes are found. Zone III is the deep zone where large rounded chondrocytes are arranged in columns perpendicular to the surface. calcified zone which rests on underlying subchondral bone.  Zone IV is the  A line called the "tide  mark" demarcates zone III from zone IV (Ghadially, 1983) (Figure 4.8B).  216  The optical  micrographs of the synovial membrane from a healthy rabbit knee joint obtained from another independent study are shown in Figure 4.9. The synovial membrane is one to three layers in thickness lying on top of the adipose tissues. A l l rabbits induced with arthritis showed cartilage destruction in both knee joints and the severity ranged from score 1 to 4 (Figure 4.10).  Fissures or clefts were observed  perpendicular to the surface of the cartilage and extended to the transitional zone (score 2) (Figure 4.10 B) or radial zone (score 3-4) (Figure 4.10 C-D) resulting in disorganization of the cellular rows and clustering of the cells in the regions.  The  disruption of the integrity of the tidemark due to the fissures and clefts (score 4) could be also observed in some rabbits (Figure 4.10 D).  The destruction of the cartilage was also  marked by the loss of proteoglycan as observed by the loss of matrix staining (Figure 4.11). There were regions in the cartilage showing interstitial loss of the matrix accompanied by the abnormality of the cells (Figure 4.1 IB and C).  It appeared that  these regions were more readily observable in the inflamed joints treated with PBS than in the joints treated with M T X microspheres.  However, it was difficult to assign scores  given each region had different sizes and different degree of loss of matrix. we did not include this feature in evaluating cartilage destruction.  o  ' 217  •  Therefore,  Pannus formation was observed in some animals from both treatment groups (Figure 4.12). Pannus is the region of proliferated synovial membrane that invades cartilage. It is usually enclosed with a membrane and attached to the cartilage. The destruction of cartilage by pannus ranged from superficial cartilage destruction (score 1) (Figure 4.12 A and B) to destruction to the depth of mid zone (score 2) (Figure 4.12 C and D). Two fold to four fold increases in the thickness of the synovial lining were observed in the animals induced with arthritis (Figure 4.13).  A few scattered and finger-like villi  with hyperplasia were also observed close to the cartilage (Figure 4.14).  Synovial  hyperplasia was accompanied by increased cellularity in the subsynovial layer indicating inflammatory cellular infiltration.  Local generation of fibrin due to inflammation could  be observed as the pink staining (Figure 4.15). Inflammatory exudates could be observed in the joints in some animals in the cavity between the synovial membrane and cartilage (Figure 4.16).  Exudates make up the  region of high protein content and cellular debris which have escaped from blood vessels and been deposited in tissues or on tissue surfaces (Ghadially, 1983).  Sometimes the  exudates may be confused with the pannus, because both exudates and pannus consist of hyperplasic synovial cells and are located in the joint cavity.  However, the exudates are  usually not enclosed with a membrane and the structure appears less dense than the  218  pannus.  The exudates were generally located closer to the synovial membrane while the  pannus was attached to the cartilage. The histological evaluation and scoring of pathological features for rabbits treated with M T X microspheres, M T X solution and PBS controls, are shown in Tables 4.4 and 4.5.  There were no significant differences in the total scores between the M T X treated  (solution or microspheres) and the PBS treated joints.  Although the total score was  higher in the knee joints treated with M T X solution than in the knee joints treated with M T X loaded microspheres, the difference was not significant. Additional micrographs demonstrating the pathological features of the synovial joints of individual animals from both treatment groups are shown in Appendix I.  219  A)  B)  Figure 4.8  Optical micrographs of the cartilage of a normal rabbit knee joint obtained  from another independent study, provided by the Faculty of Medicine, University of Calgary.  A) Cartilage of femur, tibia and part of the meniscus of the knee joint. (25x),  B).Cartilage of tibia at a higher magnification (lOOx). Zones of cartilage are marked. Round circular cells in dark purple color are the chondrocytes.  220  Figure 4.9  Optical micrographs of the synovial membrane of a normal rabbit knee joint  obtained from another independent study, provided by the Faculty of Medicine, University of Calgary.  A) Synovial membrane (25x) (indicated by the arrows), B)  Synovial membrane at a high magnification (lOOx). car: cartilage, adp: adipose cells, j cav: joint cavity.  221  Figure 4.10  Representative micrographs of cartilage destruction due to arthritis  induced by ovalbumin in rabbit knee joints. One day following the induction of arthritis, the right hind knees of the rabbits were injected with either lOmg M T X solution in 400 uL PBS or 56 mg of 18% M T X loaded PLLA2k microspheres (33-110 pm) in 400 uL PBS.  The left hind knee joints were injected with 400 pL PBS.  The rabbits were  sacrificed 14 days following M T X treatment and knee joints were processed for histological analysis. A) Minimal abrasion as indicated by the arrow on the surface of the cartilage represents score 1. The joint was injected with 400 pL PBS. B) Destruction with superficial loss of chondrocytes and cartilage disruption indicated by the arrow (score 2). The joint was injected with M T X microspheres.  C) Moderate loss  of chondrocytes and cartilage disruption to mid zone (score 3). The joint was injected with PBS.  D) Destruction with severe loss of chondrocyte to tide mark as indicated by  the arrow (score 4). The joint was injected with PBS. magnification of 25x.  A l l micrographs shown are at a  men: meniscus, car: cartilage, j cav: joint cavity.  222  A)  B)  Figure 4.11  Representative micrographs of cartilage interstitial loss of matrix due to  arthritis induced by ovalbumin in rabbit knee joints. One day following the induction of arthritis, the right hind knees of the rabbits were injected with either 10 mg M T X solution in 400uL PBS or 56 mg of 18% M T X loaded PLLA2k microspheres (33-110 um) in 400 u.L PBS.  The left hind knee joints were injected with 400 pL PBS.  The rabbits were  sacrificed 14 days following M T X treatment and knee joints were processed for histological analysis. No scores were given for this histopathological feature. interstitial matrix loss (pointed by arrows) but surface remained intact.  A)  The joint was  injected with M T X solution. B) Interstitial matrix loss accompanied by abnormal cellularity indicated by the rectangle.  The joint was injected with M T X microspheres.  C) The rectangular section of B with a higher magnification (100 x). The interstitial matrix loss was evidenced by the decrease of intensity of the color stain compared to intensity of stained cartilage in Figure 4.8. men: meniscus, car: cartilage, j cav: joint cavity.  223  A)  Figure 4.12  B)  Representative micrographs of pannus formation due to arthritis induced  by ovalbumin in rabbit knee joints. One day following the induction of arthritis, the right hind knees of the rabbits were injected with either 10 mg M T X solution in 400 uL PBS or 56 mg of 18% M T X loaded PLLA2k microspheres (33-110 um) in 400 uL PBS. left hind knee joints were injected with 400 pL PBS.  The  The rabbits were sacrificed 14  days following M T X treatment and knee joints were processed for histological analysis. A ) Pannus with superficial cartilage destruction as indicated by the rectangle represents score 1. The joint was injected with PBS. A higher magnification (lOOx) of the rectangular section is shown in B).  C) Pannus with destruction to the depth of mid zone  as indicated by the rectangle (score 2). The joint was treated with M T X microspheres. The cut parallel to the surface of the cartilage as shown by the arrow is an artifact due to histological processing. A higher magnification (lOOx) of the rectangular section is shown in D).  men: meniscus, j cav: joint cavity.  224  B)  A)  Figure 4.13  Representative micrographs of synovial tissue hyperplasia due to arthritis  induced by ovalbumin in rabbit knee joints. One day following the induction of arthritis, the right hind knees of the rabbits were injected with either 10 mg M T X solution in 400 pL PBS or 56 mg of 18% M T X loaded PLLA2k microspheres (33-110 pm) in 400 uL PBS.  The left hind knee joints were injected with 400 pL PBS.  The rabbits were  sacrificed 14 days following M T X treatment and knee joints were processed for histological analysis. A) 4-6 layer increase in synovial thickness as indicated by the arrow represents scorel. The joint was injected with PBS.  B) >7 layer increase in  synovial thickness as indicated by the double arrow (score 2). The joint was injected with PBS.  Both A) and B) are at a magnification of 25x.  225  car: cartilage.  Figure 4.14  Representative micrographs of villous hyperplasia in synovial tissues due  to arthritis induced by ovalbumin in rabbit knee joints. One day following the induction of arthritis, the right hind knees of the rabbits were injected with either 10 mg M T X solution in 400 uL PBS or 56 mg of 18% M T X loaded PLLA2k microspheres (33-110 pm) in 400 pL PBS.  The left hind knee joints were injected with 400 pL PBS.  The  rabbits were sacrificed 14 days following M T X treatment and knee joints were processed for histological analysis. A) A few scattered and short villous hyperplasia as indicated by the arrow represents score 1. The joint was injected with PBS. B) Marked finger like villi indicated by the arrow (score 2). The joint was injected with M T X microspheres. Both micrographs are at a magnification of 25 x. j cav: joint cavity.  226  A)  Figure 4.15  B)  Representative micrographs of inflammatory cellular infiltration in  synovial tissues due to arthritis induced by ovalbumin in rabbit knee joints. One day following the induction of arthritis, the right hind knees of the rabbits were injected with either 10 mg M T X solution in 400 pL PBS or 56 mg of 18% M T X loaded microspheres in 400 pL PBS.  The left hind knee joints were injected with 400 pL PBS.  The rabbits  were sacrificed 14 days following M T X treatment and knee joints were processed for histological analysis.  A) Increased cellularity and fibrin (pink stain) in the subsynovial  layer of the synovial tissue represents score 3. The joint was injected with PBS. Magnification 25x.  A higher magnification (lOOx) of the region of increased cellularity  is shown in B).  227  Figure 4.16  Representative micrographs of inflammatory exudates in the joint cavity  due to arthritis induced by ovalbumin in rabbit knee joints. One day following the induction of arthritis, the right hind knees of the rabbits were injected with either 10 mg M T X solution in 400 uL PBS or 56 mg of 18% M T X loaded microspheres in 400 uL PBS.  The left hind knee joints were injected with 400 uL PBS.  The rabbits were  sacrificed 14 days following M T X treatment and knee joints were processed for histological analysis. A ) Exudates in the joint cavity represent score 1. The joint was treated with M T X solution. (Magnification 25 x). A higher magnification (100 x) of the exudates is shown in B).  exu: inflammatory exudates, j cav: joint cavity.  228  Table 4.4  Quantitative assessment of histopathologcial features of arthritis induced by  ovalbumin in rabbit knees following M T X microspheres treatment.  One day following  the induction of arthritis, the right hind knees of the rabbits were injected with 56 mg of 18% M T X loaded PLLA2k microspheres (33-110 pm) in 400 pL PBS. knee joints were injected with 400 uL PBS. PBS in both knee joints.  The left hind  Control rabbits were injected with 400uL  The rabbits were sacrificed 14 days following M T X treatment  and knee joints were processed for histological analysis.  Histopathologcial features and scores  Treatment  Synovial  Villous  Inflammatory  Inflammatory  Pannus  Cartilage  Total  and  lining  hyperplasia  cell infiltrate  exudates  formation  destruction  score  animal #  hyperplasia  Joints injected with M T X Microspheres (n=5) #2  0  2  1  1  0  2  6  #4  2  2  2  1  1  4  12  #6  1  1  1  1  3  2  8  #10  0  1  1  1  2  4  9  #12  1  1  1  0  0  3  6  mean score  0.8+0.8  1.410.5  1.2±0.4  1±0.7  1.211.3  311  8.212.5  ±SD Contralateral joints injected with PBS (n=5) #2  0  0  0  0  3  3  #4  1  3  1  1  4  12  #6  1  1  1  1  3  8  #10  1  1  1  0  3  6  #12  1  1  0  0  4  7  0.8±0.4  1.2+1.1  0.6±0.5  0.410.5  3.410.5  7.213.3  mean score  0.8±0.8  ±SD Control animals: both joints injected with 400uL PBS (n=4) #1  0  0  0  0  0  4  4  #8  1  0  1  0  1  3  6  #1  1  0  0  0  0  4  5  #8  2  2  2  1  0  3  10  mean score  1+0.8  0.5±1  0.75±0.9  1  0.25+0.5  3.510.6  612.2  ±SD  229  Table 4.5  Quantitative assessment of histopathologcial features of arthritis induced  by ovalbumin in rabbit knees following M T X solution treatment.  One day following the  induction of arthritis, the right hind knees of the rabbits were injected with 10 mg M T X solution in 400 uL PBS.  The left hind knee joints were injected with 400 u,L PBS.  Control rabbits were injected with 400 pL PBS in both knee joints.  The rabbits were  sacrificed 14 days following M T X treatment and knee joints were processed for histological analysis. Histopathologcial features and scores  Treatment  Synovial  Villous  Inflammatory  Inflammatory  Pannus  Cartilage  Total  and  lining  hyperplasia  cell infiltrate  exudates  formation  destruction  score  animal #  hyperplasia  0  1  4  6  Joints injected with M T X solution (n=5) #3  0  0  1  #5  2  2  3  0  3  11  #7  1  1  2  1  3  9  #9  1  1  2  1  3  9  #11  2  3  3  2  4  15  mean score  1.2±0.8  1.4+1.1  2.2±0.8  1.2±0.8  1±0.7  3.4±0.5  10±3.3  0  1  4  8  ±SD  Contra lateral joints injected with PBS (n=5) #3  1  1  1  #5  1  1  2  2  1  8  #7  1  1  2  1  3  10  #9  2  1  3  1  2  10  #11  2  2  3  1  4  13  mean score  1.4±0.5  1.2±0.4  2.2±0.8  1±0.7  2.8±1.3  9.8±2.0  1±0.7  ±SD  Control animals: both joints injected with 400uL PBS (n=4) #1  0  0  0  0  4  4  #8  1  0  1  1  3  6  #1  1  0  0  0  4  5  #8  2  2  2  0  3  10  mean score  1+0.8  0.5±1  0.75±0.9  0.2510.5  3.5±0.6  6±2.2  ±SD  230  4.4 DISCUSSION  Antigen-induced arthritis in rabbits provides one of the best models of rheumatoid arthritis available (Foong and Green, 1988). The joint histopathology of antigen-induced arthritis closely resembles rheumatoid arthritis (Dumonde and Glynn, 1962; Foong and Green, 1993). It has been reported that from 1 to 4 h following intra-articular challenge in rabbits, acute inflammation associated with thick, purulent exudates, severe edema, and hemorrhagic synovitis developed and reached its peak between 24 and 48 h after antigen injection. The acute inflammatory response subsided slowly over a period of 7-10 days.  By 2-4 weeks, histological examination showed a  typical picture of chronic inflammation. The synovium appeared thickened and slightly congested and pannus tissue could be seen in some rabbits (Jasin, 1988). In this work, similar inflammatory responses were observed in the ovalbumin induced arthritis rabbit model. The knee joint swelling of all rabbits increased in the first three days following the induction of arthritis due to acute inflammation. The swelling gradually declined to similar levels as day 1 by 14 days following induction of arthritis.  Histological slides  of rabbit joints at 14 days following the induction of arthritis showed that the arthritis was well developed with distinct histopathological features such as cartilage destruction, synovial hyperplasia, pannus formation, and cellular infiltration (Figures 4.10 to 4.16).  231  In this study, M T X solution or M T X loaded microspheres were given one day following the induction of arthritis.  Early use of M T X at the development of the  arthritis would favor the suppression of the arthritis, based on several studies of the efficacy of intra-articular delivery of M T X in different animals with different onset of therapy (Foong and Green, 1993; Williams et al, 1996; Williams et al, 2001). Significant variability was observed in the M T X plasma profiles of diseased rabbits following intra-articular injection of either M T X solution or M T X microspheres (Figures 4.5 and 4.6).  The large variability observed in the plasma profiles was likely due to the  arthritis developed in the joints.  Within a day following the injection of antigen in both  knee joints of the rabbits, the development of arthritis was quite variable. The degree of swelling ranged from 2.3 to 9.8 mm in the joints injected with M T X solution, and from 2.6 mm to 6.8 mm in the joints injected with M T X microspheres (Figure 4.7 C). For rabbits injected with M T X solution, the maximum M T X plasma concentrations were approximately 0.5 pg/mL for rabbits with a joint swelling greater than 8 mm (Rabbit # 7, 9, 11), while the maximum M T X plasma concentrations were approximately 5 fold higher for rabbits with a joint swelling less than 6 mm (Rabbit # 3 and #5).  Similar  relationship between the maximum M T X concentrations in plasma and the degree of joint swelling was observed in rabbit injected with M T X microspheres, such that rabbits  232  with a higher degree of joint swelling had lower maximum M T X plasma concentrations and rabbits with decreased joint swelling showed higher maximum M T X plasma concentrations.  The maximum M T X plasma concentrations of rabbits with the highest  swelling in both treatment groups also occurred at 0.5 to 1 h later than the rabbits that had smaller joint swelling.  It has been shown that most small molecules cross the synovium  in both directions by passive diffusion, and are limited primarily by the relatively long, narrow diffusion path between synovial lining cells (Simkin and Pizzorno, 1974). Joints with larger effusions as indicated by a higher degree of swelling would have lower M T X concentrations in the synovial fluid resulting in a smaller concentration gradient across the synovial membrane.  Since an increase in joint inflammation also leads to increased  synovial thickening, the rate of transport of M T X across the synovium into the circulation may be decreased.  It has been shown that the synovial thickening and cellular  infiltration in chronic rheumatoid arthritis can cause a greater than two fold increase in average transport distance from the capillary to the joint cavity (Stevens et al., 1991). Although the development of arthritis in this study was not considered chronic at the time of the injection of M T X treatments, groups investigating similar arthritis models have shown that significant synovial proliferation took place 24 h following the induction of arthritis (Howson, et al., 1986). It is possible that slower transport of M T X across  233  inflamed synovium may have resulted in both lower maximum M T X plasma levels and increased mean absorption half-life in diseased animals (Table 4.3). However, statistical analysis of the absorption half-life data did not show any significant difference due to the large standard deviations.  Similar to previous data in healthy joints, the A U C  of the group treated with M T X solution was significantly higher than the group treated with microspheres (Table 4.3) indicating that more M T X entered systemic circulation following the injection of M T X solution than M T X microspheres. 7-OH-MTX appeared in the plasma within about 0.5 h and peaked between 1 and 2 h following intra-articular injection in both treatment groups (Figures 4.5 and 4.6) indicating the rapid conversion of M T X to its metabolite 7-OH-MTX.  Significant  individual variability was also observed in the 7-OH-MTX plasma profiles. Variable plasma M T X concentrations and variable hepatic MTX-7-hydroxylase activities among individual rabbits (Kitamura et al., 1999) likely contributed to the variability observed in the 7-OH-MTX plasma profiles. Although the exact mechanism of action of M T X in the treatment of rheumatoid arthritis has not been fully elucidated, low dose M T X in rheumatoid arthritis treatment seems to exert anti-inflammatory effects by acting at different levels of the pathophysiological cascade.  The direct anti-proliferative effects on monocytic cells  234  involved in the immune and inflammatory reactions represents the first step of intervention (Cutolo et al., 2001).  However, inhibition of both monocytic and  lymphocytic proinflammatory cytokines secreted by infiltrated macrophages and lymphocytes involved in rheumatoid synovitis seems to be the key role in the sustained anti-inflammatory actions exerted by low dose M T X (Cutolo et al, 2001).  Evidence has  shown that M T X reduced the production of proinflammatory monocytic and macrophagic cytokines such as IL-1, IL-6, and TNF-a, and increased, at least, gene expression of anti-inflammatory Th2 cytokines such as IL-4 and IL-10 with resulting anti-inflammatory effects (Cutolo et al., 2001).  Williams et al (2001) have shown that  the histological progression of antigen-induced arthritis in rats was significantly inhibited following an intra-articular injection of M T X liposomal formulation. The reduction in disease severity was accompanied by a reduction in local IL-6 and IL-1 P mRNA expression in synovial tissue extracts from the rat knee joints. Although we hypothesized that the delivery of M T X to joints via a controlled release formulation should exert an anti-inflammatory effect, the results of this in vivo efficacy study did not show that intra-articular delivery of M T X significantly improved the progression of arthritis in the antigen-induced arthritis model. Based on a system of scoring the pathological features of the disease including synovial lining and villous  235  hyperplasia, inflammatory cell infiltration and exudates, pannus formation, and cartilage destruction, there was no significant difference between M T X solution and microspheres treated groups compared to PBS (control) animals. One possible explanation for the lack of efficacy in this animal model might be the severity of the disease induced. Histological analysis of the cartilage showed that 14 days following the induction of arthritis, fissures or clefts extended to the transitional zone or radial zone of the cartilage resulting in cell death and loss of proteoglycans could be also observed (Figure 4.10). These clinical features were considered rather severe in terms of cartilage destruction and only are observed in patients with a late stage of rheumatoid arthritis development (Howson et al., 1986). Howson et al (1986) have investigated the effects of the challenge dose of ovalbumin on the joint tissues of rabbits. Arthritis was induced in both knee joints of the rabbits either with 7.5 mg ovalbumin, which is considered a dose at the upper end of the range used by most investigators or a 0.1 mg dose at the lower end of the range.  The animals were sacrificed from 1 day to 8 months post arthritis  induction in the high dose group and up to 11 months in the low dose group and progression of the disease was evaluated by histological analysis. The gross and histological changes in the knee joint in the high dose group was similar to acute cartilage necrosis with profound inhibition of chondrocyte activity in the first 2 weeks resulting in  236  widespread cell death and some survival of chondrocytes.  The low dose challenged  animals on the other hand, developed a mild arthritis with a synovitis that persisted for 11 months.  The cartilage showed irregular chondrocyte inhibition, death and degeneration  of the matrix. The authors suggested that, depending on the challenge dose used, there is a tremendous variability in the degree of arthritis produced by the antigen induced arthritis model (Howson et al., 1986). In this study, 5 mg ovalbumin was used to induce arthritis in both knee joints of the rabbits.  This dose was chosen because it is the most  commonly used dose to induce arthritis in rabbits by other investigators (Foong and Green, 1988; Foong and Green, 1993; Horisawa et al, 2002b). However, given that the cartilage was severely damaged while other histopathological features were less prominent 14 days following the antigen challenge, the dose of ovalbumin may have been too high. The arthritis model developed in this study might not fully represent the disease features of rheumatoid arthritis. It is possible that the severity of the cartilage damage might have masked the efficacy of the drug. It would be beneficial to include a positive control treatment group, such as an intravenous injection of M T X to validate the therapeutic effects of M T X in controlling the progression of disease in this animal model. In this work, the disease progression was monitored for 14 days.  Foong and Green  (1993) have shown that the effect of joint swelling and surface temperature suppression  237  was not observed until 14 days following treatment with an intra-articular injection of a 1 mg M T X liposomal formulation in inflamed rabbit joints.  Gross and histological  examination of the opened joints 56 days following arthritis induction showed a significant reduction in inflammatory changes and erosion of cartilage.  Therefore,  longer time periods for monitoring may be necessary for the observation of therapeutic effects. In summary, a severe antigen-induced inflammatory arthritis was produced in rabbits who were subsequently treated with a 10 mg dose of M T X solution or M T X microspheres.  Maximum plasma M T X concentrations seemed to correlate with the  degree of inflammation such that lower maximum M T X concentrations were produced in animals with greater joint swelling.  The data showed that there was no significant  difference between M T X solution and M T X microspheres treated groups compared to PBS (control) animals.  238  Chapter 5 SUMMARIZING DISCUSSION, CONCLUSIONS AND SUGGESTIONS FOR FUTURE WORK 5.1  S U M M A R I Z I N G DISCUSSION  M T X , an antifolate drug, has been shown to exert anti-inflammatory effects by inhibiting proliferation of cells involved in the inflammatory reaction and inhibiting both monocytic and lymphocytic proinflammatory cytokines involved in rheumatoid synovitis. Intra-articular delivery of M T X has been attempted in patients for the purpose of targeting the drug to the site of action and to minimize systemic toxic effects of the drug (Marks et ah, 1976; Bird et al., 1977). However, intra-articular injections of a M T X solution were not therapeutically effective due to the rapid clearance of M T X from the synovial joint (Wigginton et al., 1980). Previous work has shown that following the intra-articular injection of liposomal formulations of M T X in both antigen induced arthritis rabbit and rat models, increased amounts of M T X were retained in the joints and joint swelling and synovium proliferation were decreased compared to the injection of M T X solution (Foong and Green, 1998; Williams et al, 1996).  Thus, we have proposed  that intra-articular injections of a controlled release M T X delivery system should maintain an effective concentration of the drug in the joint cavity. This would be  239  particularly beneficial for patients in whom the arthritis manifests in only a limited number of joints. The goals of this work were to develop and characterize controlled release polymeric microspheres formulations of M T X and to investigate the in vivo biodistribution and efficacy of microspheres formulations following intra-articular injection in an established arthritis rabbit model.  We have proposed the following ideal  properties of the microspheres formulation for this purpose:  the formulation should be  composed of biocompatible and biodegradable polymeric materials; the formulation should possess an optimal and reproducible size distribution, and the formulation should be readily injected and sterilizable.  Furthermore, based on the design of the in vivo  pharmacokinetic and efficacy study, we also proposed an optimal M T X release lifetime of 2-3 weeks. P L L A polymers were selected as a suitable polymeric biomaterial for the microspheres formulation. M T X loaded microspheres prepared using PLLA2k were selected as the lead formulation for characterization work, based on the faster and more complete release of M T X from the PLLA2k microspheres over a period of 2-3 weeks, compared to the PLLA50k and PLLAlOOk microspheres.  Microspheres in the size  range of 33-110 pm were selected for characterization based on previous work showing  240  that microspheres in this size range were well tolerated, while smaller size microspheres induced higher levels of cellular infiltration and inflammation (Liggins et al., 2004). Solid state characterization of M T X loaded microspheres showed that M T X was likely dispersed in the microsphere matrix as a particulate dispersion. The drug release profiles of M T X loaded P L L A 2 k microspheres in the 33-110 pm size range demonstrated an extensive and rapid burst phase followed by a slow controlled release phase and approximately 95% of loaded drug was released over 2 weeks.  The initial burst phase of  M T X release was likely due to dissolution of M T X near the surface of the microspheres. The cavities and pores remaining from the loss of particulate M T X following drug release near the surface of microspheres served as a preferred pathway for drug diffusion, contributing to the rapid burst release phase. The subsequent slower release phase was controlled by a combination of drug diffusion and polymer degradation.  In vitro  degradation studies of M T X loaded PLLA2k microspheres showed 20% of total mass was lost in 40 days and the molecular weight of PLLA2k was decreased by 25 % in 28 days due to polymer degradation.  Gamma irradiation sterilization produced no adverse  effects on the microspheres and had no influence on M T X release rates. Hence the formulation could be readily sterilized prior to injection using this procedure.  241  The injection of control or M T X loaded PLLA2k microspheres in healthy rabbit joints produced a mild inflammatory response over the first three days and subsided thereafter without any significant loss of proteoglycans from cartilage.  The completion  of this characterization work demonstrated that the formulation possessed the properties felt to be ideal for the desired application, as previously discussed.  Thus, PLLA2k  microspheres in the 33-110 pm size range were selected for further pharmacokinetic and efficacy studies. To understand the biodistribution and elimination of M T X following intra-articular injection, two studies were carried out in which either M T X solution or PLLA2k M T X loaded microspheres were injected into the joints of healthy rabbits with injected M T X doses of 1.5 mg and 10 mg. The plasma M T X profiles of both studies showed that M T X solution was rapidly transferred across the synovium and entered the systemic circulation while M T X microspheres released M T X slowly in the joint cavity resulting in lower plasma M T X concentrations.  Urine data showed that during the period of study, higher  amounts of drug were excreted in the urine following the injection of M T X solution than following the injection of M T X loaded microspheres.  Synovial fluid and synovial  tissues analysis at 6 h and 24 h following intra-articular injection showed that significantly higher amounts of M T X were retained in the joints injected with M T X  242  microspheres than in the joints injected with M T X solution. Histological analysis of synovial tissues showed no evidence of phagocytosis of the microspheres and evidence suggested that microspheres were located in the adipose layer of the synovial tissues. The delivery of M T X in a controlled manner using microspheres, therefore achieved a greater retention of M T X in the joint compared to delivery of M T X as a solution. It has been shown that for a typical plasma M T X concentration profile following an oral dose of 15 mg in patients with rheumatoid arthritis, the maximum plasma M T X concentration was about 0.7 u M and concentrations decreased in a triphasic manner. Constant levels of 1 n M were maintained for several days (Hillson et al., 1997). The concentration of M T X in synovial fluid has been reported to be roughly equal to the concentration of M T X in plasma at 4 and 24 h after an oral M T X dose of 10 mg/m in 2  human subjects (Herman et al, 1989). Therefore, we estimated that a target concentration for intra-articular delivery of M T X in the joint was likely to be in the range of InM to 1 u M (0.5 ng/mL to 0.5 ug/mL). In the high dose study, the amounts of M T X detected in the joints ranged from 0.1 pg to 0.5 pg at 6 h and 24 h following intra-articular injection of M T X loaded microspheres.  Based on these data, reaching  effective concentrations of M T X in the joints seems to be achievable using M T X loaded microspheres at a dose of 10 mg M T X .  The dose of M T X given to rabbits for the  243  treatment of arthritis has ranged from 1 mg total dose to 30 mg/kg given via different routes of administration (Foong and Green, 1988; Novaes et al, 1996; Neidel et al., 1998).  A pretreatment in rabbits with a low dose of M T X at 0.25 mg/kg  intramuscularly, one week before the induction of arthritis with bovine serum albumin, has been shown to reduce intra-articular production of proinflammatory cytokines, and the intensity of protein leakage, leukocyte afflux and synovial membrane polymorphonuclear cell infiltrate (Novaes et al., 1996). A treatment with M T X at a dose of 30 mg/kg per week intramuscularly for 12 weeks was shown to reduce cartilage damage in experimental osteoarthritis in rabbits and such a high dose did not cause major adverse effects on articular cartilage proteoglycan metabolism in normal rabbits (Neidel etal, 1998). The in vivo efficacy study of intra-articular M T X loaded PLLA2k microspheres and M T X solution at a dose of 10 mg M T X was conducted using an antigen-induced arthritis rabbit model. The plasma levels of M T X at different time points following the injections were also evaluated.  A higher inter-individual variability was observed in the  M T X plasma levels of the diseased animals as compared to the healthy animals likely due to the differences in the degree of the joint swelling in diseased animals.  The rabbits  with greater joint swelling were found to have lower maximum M T X plasma  244  concentrations.  The lower maximum M T X plasma concentrations observed could be  due to a lower concentration of M T X in the synovial joint diluted by a larger volume of joint effusion leading to a smaller concentration gradient across the synovium membrane. The proliferated synovium following the induction of arthritis would also slow down the rate of transport of M T X across the synovium into the circulation. Arthritis was successfully induced in rabbits with histopathological features of synovial lining hyperplasia, villous hyperplasia, inflammatory cell infiltrate, inflammatory exudates, pannus formation and cartilage destruction clearly observed. Severe cartilage destruction was observed in all rabbits regardless of the treatments received.  Based on a system of scoring the pathological features of the disease, there  was no significant difference between M T X solution and microspheres treated groups compared to PBS (control) animals, even though evidence has shown that more drug was retained in the joint cavity of the rabbits injected with M T X microspheres compared to the rabbits treated with M T X solution. We speculate that there may be several reasons that might explain the lack of efficacy observed in this animal study.  The lack of the  efficacy in the antigen induced arthritis rabbit model might be due to the severity of the disease induced.  As discussed previously, a high level of joint inflammation including  severe cartilage destruction was observed.  Therefore, the levels of M T X achieved in the  245  joint tissues were likely not sufficient to inhibit disease progression.  It is possible that  the sustained levels of M T X released from microspheres after the burst phase was complete, were insufficient to maintain effective M T X tissue levels.  There was a high  degree of inter-individual variability observed among the animals in the same treatment group. It was apparent that the sample sizes in each treatment group were also not high enough to enable us to observe any. statistical differences between treatment groups and controls. In a similar antigen-induced arthritis rabbit model to that used in this work, an intra-articular injection of M T X liposomal formulation at lmg M T X dose was shown to reduce joint swelling and reduced inflammatory changes and erosion of cartilage. However, the reduction of joint swelling was not observed until 14 days following the intra-articular injection and the reduction in inflammatory changes and erosion was observed by histological analysis 56 days following arthritis induction (Foong and Green, 1993). It is possible that a lag time period exists before therapeutic effects may be observed following the injection of M T X treatments. Therefore, more than a 14 day time period may be necessary for monitoring the disease progression.  5.2  CONCLUSIONS  1.  M T X was successfully encapsulated in P L L A microspheres manufactured  from PLLA2k, 50k and 100k polymers. 246  2.  M T X was released from P L L A microspheres (33-110 um) with a rapid burst  phase followed by a slow controlled released phase. 3.  Control and M T X loaded PLLA2k (33-110 um) microspheres were well  tolerated in healthy rabbit knee joints for 7 days. 4.  Following an intra-articular injection in the knee joints of healthy rabbits, M T X  solution was rapidly absorbed across the synovium and entered the systemic circulation, while M T X loaded microspheres slowed down the uptake of M T X into the circulation due to the slow and controlled release of drug in the joint. 5. Significantly higher amounts of M T X were retained in the knee joints of healthy rabbits injected with M T X microspheres than in the joints injected with M T X solution. 6.  Synovial inflammation was successfully induced in the knee joints of rabbits  by ovalbumin with histopathological features resembling rheumatoid arthritis. 7.  M T X loaded microspheres or M T X solution given intra-articularly did not  significantly improve the progression of synovitis in rabbits compared to PBS treated controls.  5.3  SUGGESTIONS FOR FUTURE  W O R K  We believe it is worthwhile continuing to investigate the in vivo efficacy of M T X loaded microspheres following intra-articular injection.  247  However, it is clear from this  preliminary efficacy study that future work should incorporate a number of revisions to the protocols, as follows.  The number of animals in each treatment group should be  increased to compensate for inter-individual variability.  Different doses of antigen  should be investigated to achieve an antigen induced arthritis model with less severe cartilage destruction.  A dose escalation study should be conducted to determine the  optimal dose of M T X treatment, and a longer time period may be necessary for the observation of the progression of the disease.  For the evaluation of therapeutic effects  of M T X , histological analysis and measuring joint swelling may not be sensitive enough to reflect the pharmacological effects of M T X in a short term study.  A more  quantitative analysis such as measuring the production of proinflammatory cytokines IL-1, IL-6, or T N F - a in the synovial joints, which are known to be modulated by the administration of M T X , should also be included for a better indication of the therapeutic effects of intra-articular M T X treatments. In the pharmacokinetic study, it was difficult to retrieve M T X and microspheres from synovial fluid resulting in an underestimation of the amounts of M T X in the joints. This problem may be alleviated by using radio-labeled M T X enabling the whole joint to be homogenized for measuring total radioactivity. In this study, the slow controlled release phase following the burst phase of M T X release from microspheres may not provide sufficient amounts of M T X in the joint to  248  maintain effective M T X tissue levels. Future studies should investigate formulations of M T X loaded microspheres using other biomaterials to achieve an optimal drug release in which the rapid burst phase is reduced and M T X is released ideally at a constant rate.  249  REFERENCES Albertioni, F., Flato, B., and Seideman, P. 1995, 'Methotrexate in juvenile rheumatoid arthritis', European Journal of Clinical Pharmacology, vol. 47, pp. 507-511. Allcock, H . and Lampe, F. 1981, Contemporary Polymer Chemistry, Prentice-Hall Inc., Englewood Cliffs, New Jersey. American College of Rheumatology Subcommittee on Rheumatoid Arthritis Guidelines 2002, 'Guidelines for the management of rheumatoid arthritis: 2002 update', Arthritis and Rheumatism, vol. 46, pp. 328-346. Baker, R. 1987, Controlled Release of Biologically Active Agents, Academic press, New York. Bannwarth, B., Pehourcq, F., Schaeverbeke, T., and Dehais, J. 1996, 'Clinical pharmacokinetics of low-dose pulse methotrexate in rheumatoid arthritis', Clinical Pharmacokinetics, vol. 30, no. 3, pp. 194-210. Bergsma, E. J., Rozema, F. R., Bos, R. R., and de Bruijn, W. C. 1993, 'Foreign body reactions to resorbable poly(L-lactide) bone plates and screws used for the fixation of unstable zygomatic fractures', Journal of Oral Maxillofacial Surgery, vol. 51, no. 6, pp. 666-670. Billmayer, F. W. 1984, Polymer Science, Wiley-Interscience, New York. Bingham, C. O. 2002, 'The pathogenesis of rheumatoid arthritis: pivotal cytokines involved in bone degradation and inflammation', Journal of Rheumatology, Suppl, vol. 65, pp. 3-9. Bird, H. A., Ring, E. F., Daniel, R., and Bacon, P. A . 1977, 'Comparison of intra-articular methotrexate with intra-articular triamcinolone hexacetonide by thermography', Current Medical Research and Opinion, vol. 5, no. 2, pp. 141-146. Blanco-Prieto, M . J., Besseghir, K., Orsolini, P., Heimgartner, F., Deuschel, C , Merkle, H. P., Nam-Tran, H., and Gander, B . 1999, 'Importance of the test medium for the release kinetics of a somatostatin analogue from poly(D,L-lactide-co-glycolide) microspheres', International Journal of Pharmaceutics, vol. 184, no. 2, pp. 243-250.  250  Blyth, T., Stirling, A., Coote, J., Land, D., and Hunter, J. A . 1998, 'Injection of the rheumatoid knee: does intra-articular methotrexate or rifampicin add to the benefits of triamcinolone hexacetonide', British Journal of Rheumatology, vol. 37, no. 7, pp. 770-772. Bombardier, C , Laine, L., Reicin, A., Shapiro, D., Burgos-Vargas, R., Davis, B., Day, R., Ferraz, M . B., Hawkey, C. J., Hochberg, M . C , Kvien, T. K., and Schnitzer, T. J. 2000, 'Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. VIGOR Study Group', New England Journal of Medicine, vol. 343, no. 21, pp. 1520-1528. Bragdon, B., Bertone, A . L., Hardy, J., Simmons, E. J., and Weisbrode, S. E. 2001, 'Use of an isolated joint model to detect early changes induced by intra-articular injection of paclitaxel-impregnated polymeric microspheres', Journal of Investigative Surgery, vol. 14, no. 3, pp. 169-182. Brahn, E. 1991, 'Animal models of rheumatoid arthritis', Clinical Orthopaedics and Related Research, vol. 265, pp. 42-53. Budavari, S. 1996, The Merck Index, 12th ed, Merck Research Laboratories, White House Station, N J . Calis, S., Bozdag, S., Kas, H . S., Tuncay, M . , and Hincal, A . A . 2002, 'Influence of irradiation sterilization on poly(lactide-co-glycolide) microspheres containing anti-inflammatory drugs', Farmaco, vol. 57, no. 1, pp. 55-62. Celli, A. and Scandola, M . 1992, 'Thermal porperties and physical aging of poly(L-lactic acid)', Polymer, vol. 33, pp. 2699-2704. Cha, Y . and Pitt, C. 1989, 'The acceleration of degradation-controlled drug delivery from polyester microspheres', Journal of Controlled Release, vol. 8, no. 3, pp. 259-265. Chabner, B., Allegra, C , Ciirt, G., and Calabresi, P. 1996, 'Antineoplastic agents,' in Goodman and Gilman's the Pharmacological Basis of Therapeutics, 9th ed, J. Hardman and L. Limbird, eds., The McGraw-Hill Companies Inc., New York, pp. 1233-1287. Chan, H.-K. and Gonda, I. 1991, 'Methotrexate: existence of different types of solids', International Journal of Pharmaceutics, vol. 68, pp. 179-190.  251  Chatterji, D. C , Frazier, A . G., and Gallelli, J. F. 1978a, 'Identification and quantitation of impurities in methotrexate', Journal of Pharmaceutical Sciences, vol. 67, no. 5, pp. 622-624. Chatterji, D. C. and Gallelli, J. F. 1978b, 'Thermal and photolytic decomposition of methotrexate in aqueous solutions', Journal of Pharmaceutical Sciences, vol. 67, no. 4, pp. 526-531. Chen, M . and Chiou, W. L . 1983, 'Pharmacokinetics of methotrexate and 7-hydroxy-methotrexate in rabbits after intravenous administration', Journal of Pharmacokinetics andBiopharmaceutics, vol. 11, no. 5, pp. 499-513. Chen, M . and Chiou, W. L . 1982, 'Tissue metabolism and distribution of methotrexate in rabbits', Drug Metabolism and Disposition, vol. 10, no. 6, pp. 706-707. Chien, Y. W. 1982, Novel Drug Delivery Systems, Marcel Dekker, New York. Cociglio, M . , Hillaire-Buys, D., and Alric, C. 1995, 'Determination of methotrexate and 7-hydroxymethotrexate by liquid chromatography for routine monitoring of plasma levels', Journal of Chromatography B: Biomedical Application, vol. 674, no. 1, pp. 101-110. Collins, E. A., Bares, J., and Billmeyer, J. F. 1973, Experiments in Polymer Science, John Wiley and Sons, New York. Gonzalez, M . , Ruseckaite, R., & Cuadrado, T. 1999, "Structural changes of polylactic-acid (PLA) microspheres under hydrolytic degradation", Journal of Applied Polymer Science, vol. 71, pp. 1223-1230. Cronstein, B. N . and Merrill, J. T. 1996, 'Mechanisms of the effects of methotrexate', Bulletin of Rheumatoid Diseases, vol. 45, pp. 6-8. Cronstein, B. N . 1997, 'The mechanism of action of methotrexate', Rheumatic Diseases Clinics of North America, vol. 23, no. 4, pp. 739-755. Cutolo, M . 1999, 'Macrophages as effectors of the immunoendocrinologic interactions in autoimmune rheumatic diseases', Annals of the New York Academy Sciences, vol. 876, pp. 32-41.  252  Cutolo, M., Bisso, A . , Sulli, A . , Felli, L., Briata, M . , Pizzorni, C , and Villaggio, B. 2000, 'Antiproliferative and antiinflammatory effects of methotrexate on cultured differentiating myeloid monocytic cells (THP-1) but not on synovial macrophages from patients with rheumatoid arthritis', Journal of Rheumatology, vol. 27, no. 11, pp. 2551-2557. Cutolo, M . , Sulli, A . , Pizzorni, C , Seriolo, B., and Straub, R. H . 2001, 'Anti-inflammatory mechanisms of methotrexate in rheumatoid arthritis', Annals of Rheumatic Diseases, vol. 60, no. 8, pp. 729-735. Davies, D., Barnett, C. H., and Cochrane, W. 1962, 'Electron microscopy of articular cartilage in the young adult rabbit', Annals of Rheumatic Diseases, vol. 21, p. 11. Derendorf, H . , Mollmann, H., Gruner, A., Haack, D., and Gyselby, G. 1986, 'Pharmacokinetics and pharmacodynamics of glucocorticoid suspensions after intra-articular administration', Clinical Pharmacology and Therapeutics., vol. 39, no. 3, pp. 313-317. Derendorf, H., Mollmann, H . , Voortman, G., van den Ouweland, F. A . , van de Putte, L. B., Gevers, G., Dequeker, J., and van Vliet-Daskalopoulou, E. 1990, 'Pharmacokinetics of rimexolone after intra-articular administration', Journal of Clinical Pharmacology, vol. 30, no. 5, pp. 476-479. Donbrow, M . 1992, Microcapsules and Nanoparticles in Medicine and Pharmacy, C R C Press, Boca Raton. Dumonde, D. C. and Glynn, L. E. 1962, 'The production of arthritis in rabbits by an immunological reaction to fibrin', British Journal of Experimental Pathology, vol. 43, no. 4, pp. 373-383. Dyvik, O., Grislingaas, A . L., Tonnesen, H . H., and Karlsen, J. 1986, 'Methotrexate in infusion solutions - a stability test for the hospital pharmacy', Journal of Clinical and Hospital Pharmacy, vol. 11, no. 5, pp. 343-348. Edno, L., Bressolle, F., and Gomeni, R. 1996, 'Total and free methotrexate pharmacokinetics in rheumatoid arthritis patients', Therapeutic Drug Monitoring, vol. 18, pp. 128-134.  253  Faisant, N . , Siepmann, J., Richard, J., and Benoit, J. P. 2003, 'Mathematical modeling of drug release from bioerodible microparticles: effect of gamma-irradiation', European Journal of Pharmaceutics and Biopharmaceutics, vol. 56, no. 2, pp. 271-279. Felson, D. T., Anderson, J. J., and Meenan, R. F. 1990, 'The comparative efficacy and toxicity of second-line drugs in rheumatoid arthritis. Results of two meta-analyses', Arthritis and Rheumatism, vol. 33, no. 10, pp. 1449-1461. Firestein, G., Alvaro-Gracia, J., and Maki, R. 1990, 'Quantitative analysis of cytokine gene expression in rheumatoid arthritis', Journal of Immunology, vol. 144, pp. 3347-3353. FitzGerald, G. and Patrono, C. 2001, 'The coxibs, selective inhibitors of cyclooxygenase-2 ', New England Journal of Medicine, vol. 345, pp. 433-442. Foong, W. C. and Green, K . L . 1988, 'Retention and distribution of liposome-entrapped [3 H] methotrexate injected into normal or arthritic rabbit joints', Journal of Pharmacy and Pharmacology, vol. 40, no. 7, pp. 464-468. Foong, W. C. and Green, K . L . 1993, 'Treatment of antigen-induced arthritis in rabbits with liposome-entrapped methotrexate injected intra-articularly', Journal of Pharmacy and Pharmacology, vol. 45, no. 3, pp. 204-209. Fort, J. and Mitra, A . 1990, 'Solubility and stability characteristics of a series of methotrexate dialkyl esters', International Journal of Pharmaceutics, vol. 59, pp. 271-279. Fraser, J., Murdoch, S., Curtain, C , and Watt, B. 1977, 'Proteins retained with hyaluronic acid during ultrafiltration of synovial fluid', Connective Tissue Research, vol. 5, pp. 61. Furst, D. E. 1990, 'Clinical pharmacology of low-dose methotrexate,' in Methotrexate Therapy in Rheumatic Disease, W. S. Wilske, ed., Marcel Dekker, New York, pp. 25-33. Furst, D. E. 1995, 'Practical clinical pharmacology and drug interactions of low-dose methotrexate therapy in rheumatoid arthritis', British Journal of Rheumatology, vol. 34, Suppl 2, pp. 20-25. Furst, D. E. 1997, 'The rational use of methotrexate in rheumatoid arthritis and other rheumatic diseases', British Journal of Rheumatology, vol. 36, no. 11, pp. 1196-1204.  254  Gao, I. K., Leins, C., Bohlen, H., Heilig, B., and Lemmel, E. M . 1998, 'Inhibition of interleukin-8 synthesis by intraarticular methotrexate therapy in patients with rheumatoid arthritis', Zeitschrift fur Rheumatologie, vol. 57, no. 2, pp. 95-100. Gennaro, A . R. 1995, Remington: the Science and Practice of Pharmacy, 19th ed, Mack Publishing Company, Easton, Pennsylvania. Ghadially, F., Yong, N . , and Lalonde, J. 1982, ' A transmission electron microscopic comparison of the articular surface of cartilage processed attached to bone and detached from bone', Journal of Anatomy, Vol. 135, no. 4, pp. 685-706. Ghadially, F. 1983, Fine Structure of Synovial Joints, Butterworths, London. Gopferich, A., Alonso, M . J., and Langer, R. 1994, 'Development and characterization of microencapsulated microspheres', Pharmaceutical Research, vol. 11, no. 11, pp. 1568-1574. Gopferich, A . and Langer, S. 1995, 'Modeling of polymer erosion in three dimensions-rotationally symmertic devices', AIChE Journal, vol. 41, no. 2292, pp. 2299. Gopferich, A . 1996a, 'Polymer degradation and erosion: mechanisms and applications', European Journal of Pharmaceutics and Biopharmaceutics, vol. 42, no. 1, pp. 1-11. Gopferich, A . 1996b, 'Mechanisms of polymer degradation and erosion', Biomaterials, vol. 17, no. 2, pp. 103-114. Gossec, L. and Dougados, M . 2004, 'Intra-articular treatments in osteoarthritis: from the symptomatic to the structure modifying', Annals of the Rheumatic Diseases, vol. 63, no. 5, pp. 478-482. Goto, M . , Hosako, Y . , Katayama, M . , and Yamada, T. 1993, 'Biochemical analysis of rheumatoid synovial fluid after serial intra-articular injection of high molecular weight sodium hyaluronate', International Journal of Clinical Pharmacology Research, vol. 13, no. 3, pp. 161-166. Goto, M . , Hanyu, T., Yoshio, T., Matsuno, H., Shimizu, M . , Murata, N . , Shiozawa, S., Matsubara, T., Yamana, S., and Matsuda, T. 2001, 'Intra-articular injection of hyaluronate (SI-6601D) improves joint pain and synovial fluid prostaglandin E2 levels in rheumatoid arthritis: a multicenter clinical trial', Clinical and Experimental Rheumatology, vol. 19, no. 4, pp. 377-383. 255  Grandfils, C , Flandroy, P., and Jerome, R. 1996, 'Control of the biodegradation rate of poly(DL-lactide) microparticles intended as chemoembolization materials', Journal of Controlled Release, vol. 38, no. 2-3, pp. 109-122. Grassi, W., De Angelis, R., Lamanna, G., and Cervini, C. 1998, 'The clinical features of rheumatoid arthritis', European Journal of Radiology, vol. 27 Suppl 1, p. S18-S24. Greis, P. E., Georgescu, H . I., Fu, F. H., and Evans, C. H . 1994, 'Particle-induced synthesis of collagenase by synovial fibroblasts: an immunocytochemical study', Journal of Orthopaedic Research, vol. 12, no. 2, pp. 286-293. Grim, J., Chladek, J., and Martinkova, J. 2003, 'Pharmacokinetics and pharmacodynamics of methotrexate in non-neoplastic diseases', Clinical Pharmacokinetics, vol. 42, no. 2, pp. 139-151. Hakkarainen, M . , Albertsson, A . C , and Karlsson, S. 1996, 'Weight losses and molecular weight changes correlated with the evolution of hydroxyacids in simulated in vivo degradation of homo- and copolymers of P L A and P G A ' , Polymer Degradation and Stability, vol. 52, no. 3, pp. 283-291. Hakkarainen, M . , Karlsson, S., and Albertsson, A. C. 2000, 'Influence of low molecular weight lactic acid derivatives on degradability of polylactide', Journal of Applied Polymer Science, vol. 76, no. 2, pp. 228-239. Hansen, J., Kreilgard, B., Nielsen, O., and Veje, J. 1983, 'Kinetics of degradation of methotrexate in aqueous solution', International Journal of Pharmaceutics, vol. 16, pp. 141-152. Hasunuma, T., Kato, T., Kobata, T., and Nishioka, K. 1998, 'Molecular mechanism of immune response, synovial proliferation and apoptosis in rheumatoid arthritis', Springer Seminars in Immunopathology, vol. 20, no. 1-2, pp. 41-52. Hendel, J. and Nyfors, A . 1984, 'Pharmacokinetics of methotrexate in erythrocytes in psoriasis', European Journal of Clinical Pharmacology, vol. 27, no. 5, pp. 607-610. Herman, R. A., Veng-Pedersen, P., and Hoffman, J. 1989, 'Pharmacokinetics of low-dose methotrexate in rheumatoid arthritis patients', Journal of Pharmaceutical Sciences, vol. 78, pp. 165.  256  Hickling, P., Jacoby, R. K., and Kirwan, J. R. 1998, 'Joint destruction after glucocorticoids are withdrawn in early rheumatoid arthritis', British Journal of Rheumatology, vol. 37, no. 9, pp. 930-936. Hillson, J. L. and Furst, D. E. 1997, 'Pharmacology and pharmacokinetics of methotrexate in rheumatic disease. Practical issues in treatment and design', Rheumatic Diseases Clinics of North America, vol. 23, no. 4, pp. 757-778. Holland, S. J. and Tighe, B. J. 1992, 'Biodegradable polymers,' in Advances in Pharmaceutical Sciences, Academic Press, London, pp. 101-149. Hooper, K. A., Macon, N . D., and Kohn, J. 1998, 'Comparative histological evaluation of new tyrosine-derived polymers and poly (L-lactic acid) as a function of polymer degradation', Journal of Biomedical Materials Research, vol. 41, no. 3, pp. 443-454. Horisawa, E., Kubota, K., Tuboi, I., Sato, K., Yamamoto, H., Takeuchi, H., and Kawashima, Y . 2002a, 'Size-dependency of DL-lactide/glycolide copolymer particulates for intra-articular delivery system on phagocytosis in rat synovium', Pharmaceutical Research, vol. 19, no. 2, pp. 132-139. Horisawa, E., Hirota, T., Kawazoe, S., Yamada, J., Yamamoto, H., Takeuchi, H., and Kawashima, Y . 2002b, 'Prolonged anti-inflammatory action of DL-lactide/glycolide copolymer nanospheres containing betamethasone sodium phosphate for an intra-articular delivery system in antigen-induced arthritic rabbit', Pharmaceutical Research, vol. 19, no. 4, pp. 403-410. Howe, H . S. 1998, 'Rheumatoid arthritis: a review', Annals of the Academy of Medicine, Singapore, vol. 27, no. 1, pp. 83-88. Howson, P., Shepard, N . , and Mitchell, N . 1986, 'The antigen induced arthritis model: the relevance of the method of induction to its use as a model of human disease', Journal of Rheumatology, vol. 13, no. 2, pp. 379-390. Hunneyball, I. M . 1986, Tntra-articular administration of drugs', Pharmacy International, vol. 5, pp. 118-122. Hunter, J. A . and Blyth, T. H . 1999, ' A risk-benefit assessment of intra-articular corticosteroids in rheumatic disorders', Drug Safety, vol. 21, no. 5, pp. 353-365.  257  Iagnocco, A., Cerioni, A., Coari, G., Ossandon, A., Masciangelo, R., and Valesini, G. 2005, Tntra-articular methotrexate in the treatment of rheumatoid arthritis and psoriatic arthritis: a clinical and sonographic study', Clinical Rheumatology, Sep (E publication ahead of print) Itoh, K. 1992, Tntra-articular injection therapy in patient with rheumatoid arthritis', Nippon Rinsho, vol. 50, no. 3, pp. 531-535. Itoh, T., Ono, K., Koido, K . I., L i , Y . H., and Yamada, H . 2001, 'Stereoselectivity of the folate transporter in rabbit small intestine: studies with amethopterin enantiomers', Chirality, vol. 13, no. 3, pp. 164-169. Iven, H., Brasch, H., and Engster, J. 1985, 'Pharmacokinetics of methotrexate and 7-hydroxy-methotrexate in rabbits', Cancer Chemotherapy and Pharmacology, vol. 15, no. 2, pp. 115-120. Iven, H. and Brasch, H. 1990, 'Cephalosporins increase the renal clearance of methotrexate and 7-hydroxymethotrexate in rabbits', Cancer Chemotherapy and Pharmacology, vol. 26, no. 2, pp. 139-143. Jain, R. A . 2000, 'The manufacturing techniques of various drug loaded biodegradable poly(lactide-co-glycolide) (PLGA) devices', Biomaterials, vol. 21, no. 23, pp. 2475-2490. Jalil, R. and Nixon, J. R. 1989, 'Microencapsulation using poly(L-lactic acid). I: Microcapsule properties affected by the preparative technique', Journal of Microencapsulation, vol. 6, no. 4, pp. 473-484. Jalil, R. and Nixon, J. R. 1990, 'Biodegradable poly(lactic acid) and poly(lactide-co-glycolide) microcapsules: problems associated with preparative techniques and release properties', Journal of Microencapsulation, vol. 7, no. 3, pp. 297-325. Jasin, H. E. 1988, 'Chronic arthritis in rabbits', Methods in Enzymology, vol. 162, pp. 379-385. Jensen, L. T., Henriksen, J. H., Olesen, H. P., Risteli, J., and Lorenzen, I. 1993, 'Lymphatic clearance of synovial fluid in conscious pigs: the aminoterminal propeptide of type III procollagen', European Journal of Clinical Investigation, vol. 23, no. 12, pp. 778-784.  258  Kahn, C , Hollander, J., and Schumacher, H. 1970, 'Corticosteroid crystals in synoival fluid', The Journal of the American Medical Association, vol. 211, pp. 807-809. Kapila, S., Lee, C , Tavakkoli Jou, M . R., Miller, A . J., & Richards, D. W. 1995, 'Development and histologic characterizations of an animal model of antigen-induced arthritis of the juvenile rabbit temporomandibular joint', Journal of Dental Research, vol. 74, no. 12, pp. 1870-1879. Kirchain, W. and Gill, M . 1996, 'Drug-induced liver disease', in Pharmacotherapy, 3rd ed, J. Dipiro et al., eds., Appleton and lange, Stamford, pp. 801-814. Kirwan, J. R. 1995, 'The effect of glucocorticoids on joint destruction in rheumatoid arthritis. The Arthritis and Rheumatism Council Low-Dose Glucocorticoid Study Group', New England Journal of Medicine, vol. 333, no. 3, pp. 142-146. Kitamura, S., Sugihara, K., Nakatani, K., Ohta, S., Ohhara, T., Ninomiya, S., Green C.E., and Tyson, C. A . 1999, 'Variation of hepatic methotrexate 7-hydroxylase activity in animals and humans', IUBMB Life, vol. 48, no. 6, pp. 607-611. Kitamura, S., Nakatani, K., Sugihara, K., and Ohta, S. 1999, 'Strain differences of the ability to hydroxylate methotrexate in rats', Comparative Biochemistry and Physiology. Part C, Pharmacology, Toxicology and Endocrinology, vol. 122, no. 3, pp. 331-336. Knight, A . D. and Levick, J. R. 1983, 'The density and distribution of capillaries around a synovial cavity', Quarterly Journal of Experimental Physiology, vol. 68, pp. 629-644. Knox, P., Levick, J. R., and McDonald, J. N . 1988, 'Synovial fluid- its mass, macromolecular content and pressure in major limb joints of the rabbit', Quarterly Journal of Experimental Physiology, vol. 73, no. 1, pp. 33-45. Kremer, J., Galivan, J., and Streckfuss, A. 1986, 'Methotrexate metabolism analysis in blood and liver of rheumatoid arthritis patients: association with hepatic folate deficiency and formation of polyglutamates', Arthritis and Rheumatism, vol. 29, pp. 832-835. Kulkarni, R. K., Moore, E. G., Hegyeli, A . F., and Leonard, F. 1971, 'Biodegradable poly(lactic acid) polymers', Journal of Biomedical Material research, vol. 5, pp. 169-181. Langer, R. 1990, 'New methods of drug delivery', Science, vol. 249, pp. 1527-1533.  259  Lemperle, G., Morhenn, V . B., Pestonjamasp, V., and Gallo, R. L . 2004, 'Migration studies and histology of injectable microspheres of different sizes in mice', Plastic and Reconstructive Surgery, vol. 113, no. 5, pp. 1380-1390. Levick, J. R. 1987, 'Synovial fluid and trans-synovial flow in stationary and moving joints', in Joint Loading: Biology and Health of Articular Structures, H. Helminen et al., eds., Bristol: Wright and Sons, pp. 149-186. Levick, J. R. and McDonald, J. N . 1989, 'Ultrastructure of transport pathways in stressed synovium of the knee in anaesthetized rabbits', Journal of Physiology, vol. 419, pp. 493-508. Levick, J. R. and McDonald, J. N . 1990, 'Microfibrillar meshwork of the synovial lining and associated broad banded collagen: a clue to identity', Annals of the Rheumatic Diseases, vol. 49, no. 1, pp. 31-36. Levick, J. R. 1995, 'Microvascular architecture and exchange in synovial joints', Microcirculation, vol. 2, no. 3, pp. 217-233. Levick, J. R. and McDonald, J. N . 1995, 'Fluid movement across synovium in healthy joints: role of synovial fluid macromolecules', Annals of the Rheumatic Diseases, vol. 54, no. 5, pp. 417-423. L i , S. 1999, 'Hydrolytic degradation characteristics of aliphatic polyesters derived from lactic and glycolic acids', Journal of Biomedical Materials Research, vol. 48, no. 3, pp. 342-353. Liggins, R. 1998, Paclitaxel loadedpoly(L-lactic acid) microspheres: characterization and intraperitoneal administration, PhD thesis, University of British Columbia. Liggins, R. and Burt, H . M . 2001, 'Paclitaxel loaded poly(L-lactic acid) micospheres: properties of microspheres made with low molecular weight polymers', International Journal of Pharmaceutics, vol. 222, pp. 19-33. Liggins, R., Cruz, T., Min, W., Liang, L. S., Hunter, W. L., and Burt, H . M . 2004, Tntra-articular treatment of arthritis with microsphere formulations of paclitaxel: biocompatibility and efficacy determinations in rabbits', Inflammation Research vol. 53, pp. 363-372.  260  Lindqvist, U . , Tolmachev, V., Kairemo, K., Astrom, G., Jonsson, E., and Lundqvist, H . 2002, 'Elimination of stabilised hyaluronan from the knee joint in healthy men', Clinical Pharmacokinetics, vol. 41, no. 8, pp. 603-613. Magilavy, D. 1990, 'Animal models of chronic inflammatory arthritis', Clinical Orthopaedics and Related Research, vol. 259, pp. 38-45. Makino, K., Ohshima, H., and Kondo, T. 1986, 'Mechanism of hydrolytic degradation of poly(L-lactide) microscapsules: effects of pH, ionic strength and buffer concentration', Journal of Microencapsulation, vol. 3, pp. 203-212. Mark, J. E., Graessley, W. W., Mandelkern, L., and Koenig, J. L . 1984, Physical Properties of Polymers, American Chemical Society, Washington, D.C. Marks, J. S., Stewart, I. M . , and Hunter, J. A . 1976, 'Intra-articular methotrexate in rheumatoid arthritis', Lancet, vol. 2, no. 7990, pp. 857-858. Martinez-Sancho, C , Herrero-Vanrell, R., and Negro, S. 2004, 'Study of gamma-irradiation effects on aciclovir poly(D,L-lactic-co-glycolic) acid microspheres for intravitreal administration', Journal of Controlled Release, vol. 99, no. 1, pp. 41-52. Maxwell, S. R. and Webb, D. J. 2005, 'COX-2 selective inhibitors- important lessons learned', Lancet, vol. 365, no. 9458, pp. 449-451. McElnay, J. 1988, 'Stability of methotrexate and vinblastine in burette administration sets', International Journal of Pharmaceutics, vol. 47, pp. 236-247. McLaughlin, R . M . and Fish, R.E. 1994, 'Clinical biochemistry and hematology', in the Biology of the Laboratory Rabbit, 2 ed, P.J Manning, D.H. Ringler, and C.E New comer, eds., Academic Press,San Diego, pp. 111-127 Meachim, G. and Stockwell, R. 1979, 'The matrix', in Adult Articular Cartilage, M . A . Freeman, ed., Pitman Medical Publishing Co., London. Migliaresi, C , Fambri, L., and Conn, D. 1994, ' A study on the in vitro.degradation of polyflactic acid)', Journal of Biomaterials Science, Polymer Edition, vol. 5, no. 6, pp. 591-606. Mikuls, T. R. and O'Dell, J. 2000, 'The changing face of rheumatoid arthritis therapy: results of serial surveys', Arthritis and Rheumatism, vol. 43, no. 2, pp. 464-465. 261  Montanari, L., Costantini, M . , Signoretti, E. C , Valvo, L., Santucci, M . , Bartolomei, M . , Fattibene, P., Onori, S., Faucitano, A., Conti, B., and Genta, I. 1998, 'Gamma irradiation effects on poly(DL-lactictide-co-glycolide) microspheres', Journal of Controlled Release, vol. 56, no. 1-3, pp. 219-229. Montanari, L., Cilurzo, F., Valvo, L., Faucitano, A., Buttafava, A., Groppo, A., Genta, I., and Conti, B. 2001, 'Gamma irradiation effects on stability of poly(lactide-co-glycolide) microspheres containing clonazepam', Journal of Controlled Release, vol. 75, no. 3, pp. 317-330. Mould, A . W., Tonks, I. D., Cahill, M . M . , Pettit, A . R., Thomas, R., Hayward, N . K., & Kay, G. F. 2003, 'Vegfb gene knockout mice display reduced pathology and synovial angiogenesis in both antigen-induced and collagen-induced models of arthritis', Arthritis and Rheumatism, vol. 48, no. 9, pp. 2660-2669. Neidel, J., Schroers, B., and Sintermann, F. 1998, 'The effects of high-dose methotrexate on the development of cartilage lesions in a lapine model of osteoarthritis', Archives of Orthopaedic and Trauma Surgery, vol. 117, no. 4-5, pp. 265-269. Netter, P., Bannwarth, B., and Royer-Morrot, M . 1989, 'Recent findings on the pharmacokinetics of non-steroidal anti-inflammatory drugs in synovial fluid', Clinical Pharmacokinetics, vol. 17, no. 3, pp. 145-162. Nicholson, J. 1997, The Chemistry of Polymers, 2nd ed, The Royal Society of Chemistry, Cambridge, U . K . Nishide, M . , Kamei, S., Takakura, Y., Tamai, S., Tabata, Y., and Ikada, Y . 1999, 'Fate of biodegradable DL-lactic acid oligomer microspheres in the articulus', Journal of Bioactive and Compatible Polymers, vol. 14, pp. 385-398. Novaes, G. S., Mello, S. B., Laurindo, I. M . , and Cossermelli, W. 1996, 'Low dose methotrexate decreases intraarticular prostaglandin and interleukin 1 levels in antigen induced arthritis in rabbits', Journal of Rheumatology, vol. 23, no. 12, pp. 2092-2097. O'Dell, J. R. 2004, 'Therapeutic strategies for rheumatoid arthritis', New England Journal of Medicine, vol. 350, no. 25, pp. 2591-2602. Odian, G. 1991, Principles of Polymerization, 3rd ed, Wiley-Interscience, New York.  262  Odonnell, P. B . and Mcginity, J. W. 1997, 'Preparation of microspheres by the solvent evaporation technique', Advanced Drug Delivery Reviews, vol. 28, no. 1, pp. 25-42. Ohzawa, N . , Takahashi, Y . , Ogihara, T., Nakai, Y., and Ishiguro, J. 1997, 'Metabolic fate of ulinastatin (2); Pharmacokinetics in rabbits following intra-articular administration', Biological and Pharmaceutical Bulletin, vol. 20, no. 7, pp. 732-738. Okada, H. 1997, 'One-and three-month release injectable microspheres of the Lh-Rh superagonist leuprorelin acetate', Advanced Drug Delivery Reviews, vol. 28, no. 1, pp. 43-70. Olsen, N . J. and Stein, C. M . 2004, 'New drugs for rheumatoid arthritis', New England Journal of Medicine, vol. 350, no. 21, pp. 2167-2179. Ortendahl, M . , Holmes, T., Schettler, J. D., and Fries, J. F. 2002, 'The methotrexate therapeutic response in rheumatoid arthritis', Journal of Rheumatology, vol. 29, no. 10, pp. 2084-2091. Park, H . and Park, K . 1995, 'Biocompatibility issues of implantable drug delivery systems', Pharmaceutical Research, vol. 13, no. 12, pp. 1770-1776. Piskin, E. 1995, 'Biodegradable polymers as biomaterials', Journal of Biomaterials Science. Polymer Edition, vol. 6, no. 9, pp. 775-795. Pistner, H., Bendix, D. R., Muhling, J., and Reuther, J. F. 1993, 'Poly(L-lactide):  a  long-term degradation study in vivo. Part III. Analytical characterization', Biomaterials, vol. 14, pp. 291-298. Pitt, C. G., Gratzl, M . M . , Jeffcoat, A. R., Zweidinger, R., and Schindler 1979a, 'Sustained drug delivery systems II: Factors affecting release rates from poly(epsilon-caprolactone) and related biodegradable polyesters', Journal of Pharmaceutical Sciences, vol. 68, no. 12, pp. 1534-1538. Pitt, C. G., Jeffcoat, A . R., Zweidinger, R. A., and Schindler, A . 1979b, 'Sustained drug delivery systems. I. The permeability of poly(epsilon-caprolactone), poly(DL-lactic acid), and their copolymers', Journal of Biomedical Materials Research, vol. 13, no. 3, pp. 497-507.  263  Pitt, G. G. and Schindler, A . 1980, 'The design of controlled drug delivery systems based on biodegradable polymers', in Biodegradable and Delivery Systems for Contraception, E. Hafez and W. Van Os, eds., M T P Press Limited, Lancaster, pp. 17-45. Rahman, L. K . and Chhabra, S. R. 1988, 'The chemistry of methotrexate and its analogues', Medicinal Research Reviews, vol. 8, no. 1, pp. 95-155. Ramesh, D. V., Tabata, Y . , and Ikada, Y. 1999, 'Biodegradable microspheres for local drug release in the articulus', Journal of Bioactive and Compatible Polymers, vol. 14, pp. 137-149. Ratcliffe, J. H., Hunneyball, I. M . , Smith, A., Wilson, C. G., Davis, and SS 1984, 'Preparation and evaluation of biodegradable polymeric systems for the intra-articular delivery of drugs', Journal of Pharmacy and Pharmacology, vol. 36, no. 7, pp. 431-436. Ratcliffe, J. H., Hunneyball, I. M , Wilson, C. G., Smith, A., Davis, and SS 1987, 'Albumin microspheres for intra-articular drug delivery: investigation of their retention in normal and arthritic knee joints of rabbits', Journal of Pharmacy and Pharmacology, vol. 39, no. 4, pp. 290-295. Reitman, S, Frankel, S. 1957, ' A colorimetric method for the determination of serum glutamic oxalacetic and glutamic pyruvic transaminases. American Journal of clinical Pathology, vol. 28, no56 Robinson, J. 1997, 'Controlled drug delivery: past, present, and future', in Controlled Drug Delivery: Challenges and Strategies, K. Park, ed., American Chemical Society, Washington, D C , pp. 1-7. Romagnani, S., Matucci, A . and Rossi, O 2003, Targeted Therapies in Rheumatology eds. Smolen, J.S. and Lipsky, P.E, Martin dunitz, London-New York. Rosen, S. L . 1993, Fundamental Principles of Polymeric Materials, 2nd ed, John Wiley and Sons, Inc., New York. Salaffi, F., Manganelli, P., Carotti, M . , Subiaco, S., Lamanna, G., and Cervini, C. 1997, 'Methotrexate-induced pneumonitis in patients with rheumatoid arthritis and psoriatic arthritis: report of five cases and review of the literature', Clinical Rheumatology, vol. 16, no. 3, pp. 296-304.  264  Sasaki, K., Hosoya, R., Wang, Y . M , and Raulston, G. L. 1983, 'Formation and disposition of 7-hydroxymethotrexate in rabbits', Biochemical Pharmacology, vol. 32, no. 3, pp. 503-507. Scholes, P. D., Coombes, G. A., Davies, M . C., Ilium, L., and Davis, S. S. 1997, Controlled Drug Delivery: Challenges and Strategies, American Chemical Society, Washington, DC. Schuna, A., Schmidt, M , and Pigarelli, D. 1996, 'Rheumatoid arthritis and the seronegative spondyloarthropathies', in Pharmacotherapy: a Pathophysiologic Approach, J. Dipiro et al., eds., Appleton and Lange, Stamford, Connecticut, pp. 1717-1733. Seitz, M . , Loetscher, P., Dewald, B., Towbin, H., Rordorf, C , Gallati, H., Baggiolini, M , and Gerber, N . J. 1995, 'Methotrexate action in rheumatoid arthritis: stimulation of cytokine inhibitor and inhibition of chemokine production by peripheral blood mononuclear cells', British Journal of Rheumatology, vol. 34, no. 7, pp. 602-609. Seitz, M . , Loetscher, P., Dewald, B., Towbin, H., Rordorf, C , Gallati, Ff., and Gerber, N . J. 1996, Tnterleukin 1 (IL-1) receptor antagonist, soluble tumor necrosis factor receptors, IL-1 beta, and IL-8~markers of remission in rheumatoid arthritis during treatment with methotrexate', Journal of Rheumatology, vol. 23, no. 9, pp. 1512-1516. Senda, FL, Sakuma, E., Wada, I., Wang, H . J., Maruyama, FL, and Nobuo, M . 1999, 'Ultrastructural study of cells at the synovium-cartilage junction: response of synovial cells of the rat knee joint to intra-articularly injected latex particles', Acta Anat Nippon, vol. 74, pp. 525-535. Shargel, L. and Y u , A . 1993, Applied Biopharmaceutics and Pharmacokinetics, 3rd ed, Appletion and Lange, Norwalk, Connecticut. Shen, D. and Azarnoff, D. 1978, 'Clinical pharmacokinetics of methotrexate', Clinical Pharmacokinetics, vol. 3, pp. 1-13. Silverstein, F. E., Faich, G., Goldstein, J. L., Simon, L. S., Pincus, T., Whelton, A., Makuch, R., Eisen, G., Agrawal, N . M . , Stenson, W. F., Burr, A . M . , Zhao, W. W., Kent, J. D., Lefkowith, J. B., Verburg, K . M . , and Geis, G. S. 2000, 'Gastrointestinal toxicity with celecoxib vs nonsteroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis: the C L A S S study: A randomized controlled trial. Celecoxib Long-term Arthritis  265  Safely Study', The Journal of American Medical Association, vol. 284, no. 10, pp. 1247-1255. Simkin, P. A . and Pizzomo, J. E. 1974, 'Transynovial exchange of small molecules in normal human subjects', Journal of Applied Physiology, vol. 36, no. 5, pp. 581-587. Simkin, P. A . 1979, 'Synovial permeability in rheumatoid arthritis', Arthritis and Rheumatism, vol. 22, pp. 689-696. Simkin, P. A . and Nilson, K . 1981, 'Trans-synovial exchange of large and small molecules', Clinics in Rheumatic Diseases, vol. 7, no. 1, pp. 99-129. Simkin, P. A . and Benedict, R. 1990, 'Iodide and albumin kinetics in normal canine wrists and knees', Arthritis and Rheumatism, vol. 33, pp. 73-79. Sinko, P. and Kohn, J. 1993, 'Polymeric drug delivery systems', in Polymeric Delivery Systems: Properties and Applications, M . A . El-Nokaly, B. A . Harpetier, and D. M . Piatt, eds., American Chemical Society, Washington, D.C., pp. 18-41. Sinnett, M , Groff, G., and Raddatz, D. 1989, 'Methotrexate pharmacokinetics in patients with rheumatoid arthritis', Journal of Rheumatology, vol. 16, pp. 745-748. Siparsky, G. L., Voorhees, K . J., and Miao, F. D. 1998, 'Hydrolysis of polylactic acid (PLA) and polycaprolactone (PCL) in aqueous acetonitrile solutions: autocatalysis', Journal of Environmental Polymer Degradation, vol. 6, no. 1, pp. 31-41. Smolen, J. S., Breedveld, F. C , Eberl, G., Jones, I., Leeming, M . , Wylie, G. L., and Kirkpatrick, J. 1995, 'Validity and reliability of the twenty-eight-joint count for the assessment of rheumatoid arthritis activity', Arthritis and Rheumatism, vol. 38, no. 1, pp. 38-43. Solomon, O. and Ciuta, I. 1962, 'Determination de la viscosite intrinseque de solution de polymers par une simple determination de la viscosite', Journal of Applied Polymer Science, vol. 6, no. 24, pp. 683-686. Stevens, C. R., Blake, D. R., Merry, P., Revell, P. A., and Levick, J. R. 1991, ' A comparative study by morphometry of the microvasculature in normal and rheumatoid synovium', Arthritis and Rheumatism, vol. 34, no. 12, pp. 1508-1513.  266  Stout, R. 1993, 'Macrophage activation by T cells: cognate and non-cognate signals', Current Opinion in Immunology, vol. 5, pp. 398-403. Tanaka, N . , Sakahashi, H., Sato, E., Hirose, K., Ishima, T., and Ishii, S. 2002, Tntra-articular injection of high molecular weight hyaluronan after arthrocentesis as treatment for rheumatoid knees with joint effusion', Rheumatology International, vol. 22, no. 4, pp. 151-154. Tracy, M . A., Ward, K. L., Firouzabadian, L., Wang, Y., Dong, N . , Qian, R., and Zhang, Y . 2000, 'Factors affecting the degradation rate of poly(lactide-co-glycolide) microspheres in vivo and in vitro', Biomaterials, vol. 20, no. 11, pp. 1057-1062. van den Berg, W. B . and Bresnihan, B. 1999, 'Pathogenesis of joint damage in rheumatoid arthritis: evidence of a dominant role for interleukin-F, Best Practice and Research in Clinical Rheumatology, vol. 13, no. 4, pp. 577-597. van den Graaff, K . 1995, Human Anatomy, 4th ed, W.C.Brown Publisher, Dubuque. van Ede, A . E., Laan, R. F., Blom, H. J., De Abreu, R. A., and van de Putte, L. B. 1998, 'Methotrexate in rheumatoid arthritis: an update with focus on mechanisms involved in toxicity', Seminars in Arthritis and Rheumatism, vol. 27, no. 5, pp. 277-292. van Everdingen, A . A., Jacobs, J. W., Siewertsz Van Reesema, D. R., and Bijlsma, J. W. 2002, 'Low-dose prednisone therapy for patients with early active rheumatoid arthritis: clinical efficacy, disease-modifying properties, and side effects: a randomized, double-blind, placebo-controlled clinical trial', Annals of Internal Medicine, vol. 136, no. 1, pp. 1-12. van Krevelen, D. W. 1997, Properties of Polymers, 3rd ed, Elsevier Science B.V., Amsterdam. Verburgh, C. A . , Hart, M . H., Aarden, L. A., and Swaak, A . J. 1993, Tnterleukin-8 (IL-8) in synovial fluid of rheumatoid and nonrheumatoid joint effusions', Clinical Rheumatology, vol. 12, no. 4, pp. 494-499. Vert, M . , L i , S. M . , Spenlehauer, G., and Guerin, P. 1992, 'Bioresorbability and biocompatiblity of aliphatic polyesters', Journal of Materials Science-Materials In Medicine, vol. 3, no. 6, pp. 432-446.  267  Vert, M . , Mauduit, J., and L i , S. 1994, 'Biodegradation of P L A / G A polymers: increasing complexity', Biomaterials, vol. 15, no. 15, pp. 1209-1213. Wang, J., Ng, C. W., Win, K . Y . , Shoemakers, P., Lee, T. K., Feng, S. S., and Wang, C. H . 2003, 'Release of paclitaxel from polylactide-co-glycolide (PLGA) microparticles and discs under irradiation', Journal of Microencapsulation, vol. 20, no. 3, pp. 317-327. Wang, Y . X . , Robertson, J. L., Spillman, W. B., Jr., and Claus, R. O. 2004, 'Effects of the chemical structure and the surface properties of polymeric biomaterials on their biocompatibility', Pharmaceutical Research, vol. 21, no. 8, pp. 1362-1373. Watts, P. J., Davies, M . C., and Melia, C. D. 1990, 'Microencapsulation using emulsification/solvent evaporation: an overview of techniques and applications', Critical Reviews in Therapeutic Drug Carrier Systems, vol. 7, no. 3, pp. 235-259. Wigginton, S., Chu, B., Weisman, M . , and Howell, S. 1980, 'Methotrexate pharmacokinetics after intraarticular injection in patients with rheumatoid arthritis', Arthritis and Rheumatism, vol. 23, no. 1, pp. 119-122. Williams, A . S., Camilleri, J. P., Goodfellow, R. M . , and Williams, B . D. 1996, ' A single intra-articular injection of liposomally conjugated methotrexate suppresses joint inflammation in rat antigen-induced arthritis', British Journal of Rheumatology, vol. 35, no. 8, pp. 719-724. Williams, A . S., Jones, S. G., Goodfellow, R. M . , Amos, N . , and Williams, B. D. 1999, Tnterleukin-lbeta (IL-lbeta) inhibition: a possible mechanism for the anti-inflammatory potency of liposomally conjugated methotrexate formulations in arthritis', British Journal of Pharmacology, vol. 128, no. 1, pp. 234-240. Williams, A . S., Topley, N . , Dojcinov, S., Richards, P. J., and Williams, B. D. 2001, 'Amelioration of rat antigen-induced arthritis by liposomally conjugated methotrexate is accompanied by down-regulation of cytokine mRNA expression', Rheumatology, vol. 40, no. 4, pp. 375-383. Woodward, S., Brewer, P., Moatmed, F., Schindler, A., and Pitt, C. 1985, 'The intracellular degradation of poly(e-caprolactone)', Journal of Biomedical Materials Research, vol. 19, pp. 437-444.  268  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0092780/manifest

Comment

Related Items