|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 SPECTRAL IMAGING BASED IN VIVO MODEL SYSTEM FOR CHARACTERIZATION OF TUMOR MICROVESSEL RESPONSE TO VASCULAR TARGETING AGENTS By MAMTA WANKHEDE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010
2 2010 Mamta Wankhede
3 To my Mom
4 ACKNOWLEDGMENTS I hereby take th e opportunity to thank all who have been a part of this journey with me. My heartiest thanks go to Dr. Brian Sorg, who has been my mentor, guide, and source of inspiration for the past five years. I can only hope that his discipline and hardworking nature has rubbed off on us as himself. I am also thankful to my committee members Dr. Dietmar Siemann, Dr. Brandi Ormerod, and Dr. Paul Carney for guiding and shaping my research in the right direction. T his work w ould not have been possible without the contribution of our collabora tors Dr. Mohan Raizada, Dr. Rodrigo Fraga Silva, Dr. Lori Rice, Chris Pampo, Sharon Lepler, Dr. Suk Paul Oh, and Dr. Abhinav Acharya. My special thanks to Chris Pampo for being there for me when I needed help the most. I take this opportunity to extend my true appreciation and gratitude for the help and contribution of all my colleagues Casey deDeugd, Se woon Choe, Jennifer Lee and Gizmo Raymond Kozikowsky Phillip Barish, and all members of Kesorgle lab toward the completion of the presented resear ch. Three cheers to all our successes and failures. Last but not the least, I would like to thank all my friends and family Vaishali, Suresh, Chandrakiran, Ravikant, Kevin, A rnav, Bhajji, Mithu, Ziggy, Isabella and Mannu for their support and patienc e through these years.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 TABLE OF CONTENTS ................................ ................................ ................................ .. 5 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF TABLES ................................ ................................ ................................ .......... 12 ABSTRACT ................................ ................................ ................................ ................... 13 CHAPTER 1 INTRODUCTI ON ................................ ................................ ................................ .... 14 Tumor versus Normal Vasculature ................................ ................................ ......... 14 Vascular Targeting for Tumor Therapy ................................ ................................ ... 15 Vascular Disrupting Agents and Anti angiogenic Agents ................................ ........ 16 OXi4503 and Other Vascular Disrupting Agents ................................ .............. 16 Anti V EGF Angiogenesis Inhibitor Avastin ................................ ....................... 19 Need for Combinatorial Approach ................................ ................................ ........... 20 Evaluation Tools for Vascular Targeting Therapy ................................ ................... 22 Motivation and Goal of Research ................................ ................................ ............ 23 2 MOUSE MODEL AND SPECTRAL IMAGING SYSTEM: AN INTRODUCTION ..... 25 Window Chamber Mouse Model: Background ................................ ........................ 25 Spectral Imaging: Background ................................ ................................ ................ 27 Imaging Acquisition and I mage Processing: Techniques Used in the Presented Research ................................ ................................ ................................ ............. 29 Imaging System ................................ ................................ ................................ 29 Image Acquisition ................................ ................................ ............................. 30 Image Processing ................................ ................................ ............................. 31 3 SPECTRAL IMAGING OF WINDOW CHAMBERS: A VERSATILE TOOL FOR VASCULAR PATHOLOGY STUDIES ................................ ................................ ..... 34 Introduction ................................ ................................ ................................ ............. 34 Materials and Methods ................................ ................................ ............................ 36 Methods for Tumor Vessel Acute Oxygenation Fluctuations ............................ 36 Preparation of fluorescent red blood cells ................................ .................. 36 Preparation of 4TO7 tumor cells ................................ ................................ 37 Window surgery, in vivo tumor initiation and tumor imaging ...................... 38 Image and data processing ................................ ................................ ........ 38
6 Red blood cell flux imaging ................................ ................................ ........ 39 Methods for Induced Thrombus Experiments ................................ ................... 40 Experiment design summary ................................ ................................ ...... 40 Surgery ................................ ................................ ................................ ...... 41 Platelet labeling and visualization ................................ .............................. 41 Thrombus induction ................................ ................................ ................... 42 Thrombus imaging ................................ ................................ ..................... 42 Methods for Arterio venous Malformation Experiments ................................ .... 42 Experiment design summary ................................ ................................ ...... 42 Window chamber wound model ................................ ................................ 43 Spectral imaging ................................ ................................ ........................ 43 Blood flow imaging ................................ ................................ ..................... 43 Results ................................ ................................ ................................ .................... 44 Tumor Microvascular Acute Oxygenation Fluctuations ................................ .... 44 Spectral Imaging for Study of Microvascular Occlusions ................................ .. 50 Spectral Imaging for Real Time Imaging of Arterio Venous Malformations ...... 53 Discussion ................................ ................................ ................................ .............. 58 4 OXi4503 TREATMENT OF 4T1 and CAKI 1 TUMORS ................................ .......... 61 Introduction ................................ ................................ ................................ ............. 61 Materials and Methods ................................ ................................ ............................ 63 Tumor Cells ................................ ................................ ................................ ...... 63 Animal Model ................................ ................................ ................................ .... 63 Drug Preparation and Experimental Design ................................ ..................... 64 Imaging System, Image Acquisition and Image Processing ............................. 64 Results ................................ ................................ ................................ .................... 65 Treatment of 4T1 Mouse Mammary Carcinomas ................................ ............. 65 Treatment of Caki 1 Human Renal Carcinomas ................................ ............... 68 Rapid Oxygenation Changes after OXi4503 Treatment ................................ ... 70 Discussion ................................ ................................ ................................ .............. 77 5 OXI4503 AND AVASTIN COMBINATION T REATMENT OF CAKI 1 TUMORS ..... 81 Introduction ................................ ................................ ................................ ............. 81 Materials and Methods ................................ ................................ ............................ 84 Tumor Cells ................................ ................................ ................................ ...... 84 Animal Model ................................ ................................ ................................ .... 84 Drug Preparation and Experimental Design ................................ ..................... 85 Vascular disrupting agent OXi4503 ................................ ............................ 85 Angiogenesis inhibitor Avastin ................................ ................................ ... 85 Individual and combinatorial treatment d esign and imaging schedule ....... 85 Tumor Size Measurement ................................ ................................ ................ 86 Imaging System, Image Acquisition and Image Processing ............................. 86 Statistical Analysis ................................ ................................ ............................ 87 Results ................................ ................................ ................................ .................... 87 Tumor Model Identification ................................ ................................ ............... 87
7 Caki 1 tumor growth after individual treatment with OXi4503 .................... 87 Effect of OXi4503 treatment on Caki 1 tumors: Hemoglobin saturation (HbSat) as sessment ................................ ................................ ............... 89 Combination Treatment with OXi4503 and Avastin ................................ .......... 91 Tumor growth study ................................ ................................ ................... 91 Quantitative HbSat analysis and comparison ................................ ............. 93 Discussion ................................ ................................ ................................ .............. 98 6 CONCLUSION AND FUTURE DIRECTION ................................ ......................... 103 LIST OF REFERENCES ................................ ................................ ............................. 105 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 122
8 LIST OF FIGURES Figure page 2 1 The window chamber model. Figure 2 1a shows the window chamber implanted on a healthy nude mouse. Figure 2 1b demonstrates the 4T1 tumor growing at the implantation site in the window chamber indicated by t he dashed circle, several days after implantation. ................................ ............. 26 2 2 Description of a spectral image data set . As described by Garini et al. each point in the 3 D spectral image cube represents a single number and the spectral image is described as I(x,y,k). It can be viewed either as an image I(x,y) at each wavelength k, or as a spectrum I(k) at every pixel (x,y). .... 28 2 3 The Imaging system. The Zeiss based imaging microscope system is shown in Figure2 2a. Figure 2 2b gives a close view of the mouse being imaged, placed on the heating pad and breathing through the nose cone. ...................... 30 2 4 Flow diagram depicting the image processing method to obtain hemoglobin saturation map from the stack of 16 images obtained via spectral imaging. ....... 33 3 1 Transmitted light image of the tumor microvessel network. It is a 4TO7 mouse mammary carcinoma with a diameter of 1.5 to 2.0 mm 9 days after implantation of cells. White triangles represent regions of interest chosen for analysis. A 10 objective was used for imaging (NA of 0.3, working distance of 5.5 mm). ................................ ................................ ................................ ......... 46 3 2 Hemoglobin saturation image of the tumor microvessel network in Figure 3 1. The pixels are colored according to the hemoglobin saturation scale to the rig ht of the figure. The background is black. Regions of interest for analysis are indicated in the figure. ................................ ................................ .................. 47 3 3 Fluorescence image of RBCs labeled with DiD flowing in the tumor microvessel n etwork. Images were captured with an electron multiplying CCD camera at video rates (30 Hz). 20 s of data were taken every minute of the hour long imaging session, and cells were counted over this time increment and then converted to RBC flux measurements. The locations of the regions of interest for analysis are indicated in the figure. ............................ 48 3 4 Plot of RBC flux versus hemoglobin saturation (HbSat) for region of interest 1 in Figure 3 1. The d ata points were acquired at 1 min intervals for 1 h. ............. 49 3 5 Plot of RBC flux versus hemoglobin saturation (HbSat) for region of interest 3 in Figure 3 1. The data points were acquired at 1 m in intervals for 1 h. ............. 49 3 6 Brightfield and HbSat images are shown before and 21 minutes after topical application of FeCl3 to induce thrombus formation. The red dotted line in the top left image indicates where the border of the paper soaked with FeCl3
9 was relative to the image area. The yellow box indicates the region shown in the images in Figure 3 7 at higher magnification. ................................ ............... 51 3 7 Brightfield and HbSat time sequence images of thrombus development for the region indicated by the yellow box in Figure 3 6. The yellow outlined area on the venule in the top left image indicates the ROI used for the HbSat measurements shown in Fi gure 3 8. The arrow in the brightfield image at the 8 min time point indicates a forming thrombus. ................................ .................. 52 3 8 Plot of the HbSat in the ROI indicated in Figure 4 2 at various time points during t hrombus development. The data points in the figure are given as the medianinterquartile range for the ROI. ................................ ............................. 53 3 9 AVM development in a conditional Alk1 deletion mouse (ROSA26 CreER /Alk1 stimulated by a wound, where A, arteriole; V, venule; and W, wound. The left column shows transmitted light images and the right column shows HbSat images. Region where AVMs formed is indicated in the day 5 and day 9 images. At day 5, the AVM connect ions were not obvious but clear changes in venule oxygenation could be seen due to the AVMs. The arrowheads indicating the arteriole and venule also indicate regions of interest where HbSat measurements were obtained. ................................ ......... 56 3 10 Wound healing in a control mouse: A, arteriole; V, venule; and W, wound. The left column shows transmitted light brightfield images and the right column shows HbSat images. The arrowheads indicating the arteriole and ven ule also indicate regions of interest where HbSat measurements were obtained. Unlike the conditional Alk1 deletion mouse in Figure 3 9, no AVMs formed during wound healing. ................................ ................................ ............ 57 4 1 Typical OX i4503 treatment progression in a 4T1 tumor bearing mouse. The brightfield images (left hand figures in each panel) show structural alterations in the vasculature through the course of the single treatment. The pproximate tumor border is indicated by the do tted line in (A). Vessels outside the dotted line are considered normal tissue vasculature. The tumor periphery consisted of a 250 to 500 Remaining tumor vasculature except for the periphery constituted th e tumor core. Images were obtained at x2.5 magnification with image dimensions of 4.15x3.125 mm. The HbSat maps are shown in the right hand figures in each panel. The oxygenation levels in the HbSat maps are color coded as indicated by the colorbar. Arrows in frames (G) and (H) highlight the disintegrating vasculature, while arrows in (I) and (J) indicate an example vessel that persists through all time points with only oxygenation changes. The star in frames (M) and (O) indicate the avascular regions creat ed by the OXi4503 treatment. ................................ ................................ ............................ 71 4 2 Recovery, re oxygenation, vascular remodeling, and neovascularization in the 4T1 tumor during the recovery phase after OXi4503 treatment for the
10 area indi cated in Figure 1O by the dotted rectangle. The approximate tumor border is demarcated by the dotted line in the first bright field image. The arrows highlight an example vessel that maintains its structure throughout the treatment but undergoes dramatic o xygenation changes. The color scale for hemoglobin saturation is the same as in Figure 4 1. ................................ ..... 72 4 3 Quantitative tumor regional analysis for 4T1 tumor treatment. The tumor was divided into co re, periphery, and normal tissue for purposes of analysis. (A) Brightfield images before and after every successive treatment of the tumor in Figure 4 1. (B) Cumulative HbSat values (mean standard error) for 3 serial treatments in all 5 4T1 tumor mice. A rrows indicate drug administration. (C) Cumulative HbSat values (mean standard error) for 3 serial sham treatments (saline) in all 3 control mice. ................................ .......... 73 4 4 OXi4503 treatment 2 of a Caki 1 tu mor bearing mouse. Images were obtained at x2.5 magnification with image dimensions of 4.15x3.125 mm. The HbSat maps are shown in the right hand figures in each panel. The oxygenation levels in the HbSat maps are color coded as indicated by the colorbar. ................................ ................................ ................................ ............. 74 4 5 Quantitative tumor regional analysis for Caki 1 tumor treatment. The tumor was divided into core, periphery, and normal tissue for purposes of analysis. (A) Brightfield images before a nd after every successive treatment of the tumor in Figure 6 4. (B) Cumulative HbSat values (mean standard error) for 3 serial treatments in all 5 4T1 tumor mice. Arrows indicate drug administration. (C) Cumulative HbSat values (mean standard error) fo r 3 serial sham treatments (saline) in a control mouse. ................................ ........... 75 4 6 Rapid oxygenation changes in 4T1 tumor microvasculature after treatment with VDA OXi4503. The mouse was imaged for 20 min befor e treatment and for 60 min after treatment. Spectral images were obtained every 30 sec. (A) Brightfield image of the tumor microvasculature. (B) Hemoglobin saturation before and after OXi4503 treatment. ................................ ................................ .. 76 5 1 Effect of OXi4503 treatment on different sized Caki 1 tumors. As shown in Figure 5 1A (i) and (ii), tumors with pre OXi4503 volumes up to 50 mm 3 demonstrated control of tumor growth compared to controls in the window chamber model. ................................ ................................ ................................ .. 88 5 2 Example of one of the serial OXi4503 treatments in one mouse. The brightfield images are shown along with the corresponding hemoglobin saturation maps. At the beginning of treatment the t umor was approximately 55 mm 3 in volume and 12 days old. Structural deterioration of tumor vessel network is evident in the brightfield images up to 8 h post Tx, with recovery of peripheral vessels from 24 h. Oxygenation plummeted corresponding to the stru ctural damage nearly to 8 h and recovered on the tumor rim from 8 h. At 48 h post Tx tumor had shrunk to nearly 42 mm 3 in volume. .............................. 90
11 5 3 Tumor volume comparison after individual and combinat ion treatment with OXi4503 and Avastin. It was observed that combination of therapeutics have additive/supra additive effects on tumor size reduction of Caki1 tumors. A) The carrier treatment had no effect on tumor growth retardation. B) Avastin alone did not reduce the tumor size appreciably. C) The treatment of OXi4503 alone retarded the tumor growth D) Trend suggests that the combinatorial treatment of Avastin and OXi4503 reduces the tumor size with each consecutive treatment. E) Interesting significantl y different pairs (p<0.05) are as follows: treatments marked with as compared to 48h post Tx2 and Tx3 of Avastin. ................................ ................................ ...................... 92 5 4 Example of classification of tumor core, periphery and normal vascu lature. The periphery was chosen between 250 500 m on the tumor rim. ................... 94 5 5 The HbSat or oxygenation of the periphery, core and normal microvessels modulates with individual treatments with (A) Oxi4503 or (B) Avastin, or their (C) combina tion at different time points and (D) control. Two way ANOVA was performed on HbSat values normalized to 0 h time points of each treatment obtained from more than 3 mice per group and the average values of HbSat and standard error are plotted against corre sponding time points. Significant differences between individual time points are indicated. The black arrows represent the time point where treatment was given to mice. ........ 97
12 LIST OF TABLES Table page 1 1 Small molecule VDAs in active clinical development. Adopted from .............. 17 3 1 Comparison of arteriole and venule saturation in a mouse model of HHT during wound healing. ................................ ................................ ........................ 58
13 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SPECTRA L IMAGING BASED MODEL IN VIVO SYSTEM FOR CHARACTERIZATION OF TUMOR MICROVESSEL RESPONSE TO VASCULAR TARGETING AGENTS By Mamta Wankhede December 2010 Chair: Brian Sorg Major: Biomedical Engineering Functional vasculature is vital for tumor growth, prol iferation, and metastasis. Many tumor specific vascular targeting agents (VTAs) aim to destroy this essential tumor vasculature to induce indirect tumor cell death via oxygen and nutrition deprivation. The tumor angiogenesis inhibiting anti angiogenics (AI s) and the established tumor vessel targeting vascular disrupting agents (VDAs) are the two major players in the vascular targeting field. Combination of VTAs with conventional therapies or with each other, have been shown to have additive or supra additiv e effects on tumor control and treatment. Pathophysiological changes post VTA treatment in terms of structural and vessel function changes are important parameters to characterize the treatment efficacy. Despite the abundance of information regarding these parameters acquired using various techniques, there remains a need for a quantitative, real time, and direct observation of these phenomenon in live ani mals. Through this research we aspired to develop a spectral imaging based mouse tumor system for real time in vivo microvessel structure and functional measurements for VTA characterization. A model tumor system for window chamber studies was identified, and then combinatorial effects of VDA and AI were characterized in model tumor system.
14 CHAPTER 1 INTRO DUCTION Tumor v ersu s Normal Vasculature Under normal physiological conditions, angiogenesis or sprouting of new vessels is a tightly controlled mechanism with the pro angiogenic factors delicately balancing the anti angiogenic factors  Such p hysiological angiogenesis is required during fetal growth, wound healing process as well as in female reproductive system. The resulting vasculature is uniformly structured and well oxygenated As opposed to normal angiogenesis, pathological conditions such as tumors cause an imbalance between pro and anti angiogenic factors in favor of the pro angiogenic side, resulting in accelerated vessel growth. Such enhanced vascular development is initiated by tumors to fulfill the excessive oxygen and nutrients requirements of their rapidly growing tumor mass. The resulting t umor vasculature greatly differs from normal vasculature in both structural and molecular aspects. In most big tumors the core remains comparatively non vascularized. Vasculature in the remaining tumor areas is often found to be leaky, unorganized, and deformed resulting in a vasculature system which is maximally stimulated yet minimally fulfilling the metabol ic needs of the host tumor mass  Other vascular abnormalities include temporary occlusions, rapidly dividing endothelial cells, blind ends, and a reduction in pericytes  This results in heterogeneously distributed hostile micro milieus within the tumor mass characterized by parameters such as hypoxia, hypoglycemia, extra cellular acidosis etc. These hostile parameters may directly or indirectly be responsible for inducing angiogenesis, selection of aggressive cells, tumor progression and acquired treatment resistance.
15 I t has been observed that in order to initiat e a ggressive vessel development, tumors rely on n umerous triggering mechanisms such hypoxia, hypoglycemia, oncogenic transformations, or autocrine growth factor loops  For example hypoxia, or oxygen deprivation accelerates angiogenesis in tumors via increase in the hypoxia i nducible transcription factor (HIF 1 ), which in turn up regulates the vascular endothelial growth factor (VEGF). VEGF is considered among one of the most important regulators of angiogenes is. On the other hand, hypoglycemia or n utrient deprivation induces VEGF expression through a pathway completely independent of HIF. Growing tumor mass is commonly inflicted with hypoxic and hypoglycemic insults due to mismatch in demand and supply rate. Such pathological a ngiogenesis can thus be called as inherent characteristics of the solid tumors. Vascular Targeting for Tumor Therapy Despite its abnormal structure and function, active tumor vasculature remains crucial for tumor growth, proliferation a nd metastasis  Tumor vasculature plays essential role in tumor survival, tumor metastasis and progression, and in controlling the tumor microenvironment, thereby influencing conventional therapies  Targeting tumor vasculature could thus have devastating effects on the tumor. When compromised, impaire d tumor vasculature may result in indirect tumor cell death via nutrient deprivation and impaired waste product drainage Vascular targeting thus makes a promising therapeutic target for tumor treatment and strategies targeting tumor blood vessels have bee n actively developed over the years [7 11] The idea of exploiting the tumor vasculature as a form of treatment has been ar ound since mid 1800s, with n umerous strategies including bacterial inductions, use o f lead colloids and arsenicals, and colchicines implemented through 1900s. A radical
16 approach took roots d when Juliana Denekamp proposed to use the specificity of tumor endothelial cells as a targeting approach. She suggested that the much hi gher proliferation rate of tumor endothelial cells compared to normal tissue could be specifically targeted to destroy the tumor vasculature. Recently this approach has gained momentum and a number of different agents are being developed selectively and sp ecifically targeting the tumor vasculature without damaging the normal vasculature [5, 12 16] Vascular Disrupting A gents and Anti angiogenic Agents The vascular disrupting agents (VDAs) and angiogenic inhibitors (AIs) are two major types of vascular targeting agents currently being investigated for tumor therapy with distinct mechanisms of action, disparate treatment regimens and unique treatment application and outcomes  While vas cular disrupting agents (VDAs) utilize the inherent differences between normal and tumor vasculature to destroy the well established tumor vessels [5, 12 16] the angiogenesis inhibitors' (AIs) primary function is to inhibit new vessel formation by interfering with the key signaling pathways essential fo r neo angiogenesis [9, 10, 18 21] It has been hypothesized that these two distinct and non overlapping therapeutic approac hes can be used in a synergistic manner to achieve additive therapeutic benefit, either in combination with regular chemotherapy or radiation, or in combination with each other [6, 22 26] OXi4503 and Other Vascular Disrupting Agent s Vascular disrupting agents essentially refer to all the agents which aim at destroying the well established tumor microvasculature without damaging the normal vessels. These agents range from physical treatments such as hyperthermia or
17 photodynamic therapy (PTD), to induction of cytokines such as tumor necrotic factor (TNF) and interleukins, chemotherapeutics such as Vinca alkaloids and arsenic trioxide (ATO), antibodies, peptides, and growth factors that selectively bind to tumor vessels and ultimately destroy them  Small molecule vascular targeting agents are rapidly becoming the most popular VDAs [ 6] As evident from Table 1 1, majority of VDAs in clinical trials use inhibition of tubulin assembly as their mechanism of action  Table 1 1. Small molecule VDAs in active clinical development. Adopted from  Compound Mechanism Current development status Tumor blood flow reductions in patients Company CA4P (Zybrestat) Tubulin depolymerizing agent Phase II and III clinical trials ongoing Yes OXiGENE, Inc. DMXAA (ASA404) TNF induction Phase II clin ical trials completed Yes Antisoma/Nov artis MN 029 (Denebulin) Tubulin depolymerizing agent Phase I clinical trials completed Yes Medicinova AVE8062 (AC7700) Tubulin depolymerizing agent Phase I clinical trials ongoing Yes Sanofi Aventis ZD6126 Tubulin depolymerizing agent Phase I clinical trials completed Yes Angiogene Pharmaceutica ls Ltd OXi4503 Tubulin depolymerizing agent Phase I clinical trials ongoing Studies ongoing OXiGENE, Inc NPI2358 Tubulin depolymerizing agent Phase I clinical trials ongoin g Studies ongoing Nereus Pharmaceutica ls CYT997 Tubulin depolymerizing agent Phase I clinical trials ongoing Studies ongoing Cytopia MPC 6827 Tubulin depolymerizing agent Phase I clinical trials ongoing Studies ongoing Myriad Pharmaceutica ls There are two major sub categories of small molecule VDAs. The first sub category utilizes the cytokine induction capacity of agents like flavone acetic acid (FAA)
18 and its derivative 5, 6 dimethylxanthenone 4 acetic acid (DMXAA). The second sub category utilizes the tubulin binding capability of agents such as Combretastatin A4 Phosphate (CA4P), the phosphate prodrug of N acetyl colchinol (ZD6126), Ave8062, NPI2358, MN 029, and OXi4503. OXi4503 is the diphosphate prodrug of Combretastatin A1 (CA1), which along with i ts predecessor Combretastatin A4 Phosphate (CA4P) is isolated from the South African tree Combretum Caffrum and is considered to be among the most potent tubulin binding agents  When used as a single agent against the murine breast adenocarcinoma C aNT the murine colon tumor MAC29, the human adenocarcinoma MDA MB 231 and the murine myocardial endothelioma MHEC5 T, OXi4503 exhibits a stronger antivascular and antitumor effect than CA4P  Oxi4503 has been shown to retard tumor growth in a dose dependent manner and improved survival in murine model of colorectal liver metastases  CA4P, the analogue of OXi4503, has been shown to enhance the effects of radiation, hyperthermia, chemotherapy, and radio immunotherapy  Studies have shown that this second generation vascular disrupting agent OXi4503 not only possesses significant anti vascular effects in solid tumors but also follows a similar treatment response trend after subsequent repeat treatments  tubulin subunits inside the dividing endothelial cells, thereby preventing the formation of microtubules. The cytoskeleton of the proliferating endothelial cells gets disrupted which in turn results in endothelial cell shape changes. The endothelial cells round up, detach from the vascular wall, and finally the vascular wall collapses. This vascular collapse and the subsequent thrombus
19 formation blocks the vessel blood supply resulting in tumor cell death and extensive necrosis due to lack of oxygen, nutrition, and metabolic waste removal [3, 29] While experimental evidence has confirmed the selective tumor vessel targeting and i is still not overcome even after using the maximum tolerable doses (MTD) of the drug  Since the peripheral tumor cells obtain nutrition mainly from the surrounding normal vessels, these few peripheral tumor cells survive resul ting in incomplete tumor destruction. Consequently, a combination therapy of OXi4503 with traditional anticancer therapies is suggested for complete tumor eradication  Anti VEGF Angiogenesis Inhibitor Avastin The idea that tumor angiogenesis was mediated by diffusible factors expressed by tumor cells  and that anti angiogenic strategies might be effective against tumors  angiogenesi s inducing, and possible anti angiogenic factors, such as epidermal growth factor (EGF), transforming growth factor (TNF) TGF tumor necrosos factor and angiotensin have been reported  The vascular endothelial growth factor (VEGF also known as VEGF A ) is a key regulator of angiogenesis and the role of the VEGF gene family in the regulation of angiogenesis has been kee nly studied  While the complex process of assembly and maturation of the vessel wall is co ordinated by several factors such as angiopoietins, platelet derived growth factor B (PDGF B) etc  VEGF A is responsible for a rate limiting step in normal and pathologi cal blood vessel growth  Studies have reported VEGF expression in many human tumors including lung  breast  gastrointestinal tract  renal  and ovarian carcinomas 
20 Renal cell carcinomas have been reported to have particu larly high VEGF expressions. A mouse anti human VEGF monoclonal antibody known as A.4.6.1 has been known to possess tumor growth inhibition potential several mouse tumor models [41 44] Avastin or Bevacizumab is the humanized form of the mouse anti VEGF monoclonal antibody A.4.6.1. Similar to the mouse antibody, Avastin binds and neutralizes all human VEGF A isoforms, but does not neutralize other members of the VEGF gene family [35, 45] The pharmacokinetic propertie s of Avastin have been studied and its terminal half life has been reported as 17 21 days  After extensive preclinical [45, 47, 48] and clinical studies [49 53] Avastin has been approved by FDA for clinical use as a first line therapy for metastatic colorectal cancer. Need for Combinatorial Approach Despite their efficacy in terms of vascular damage and tumor cell kill, the VDAs suffer from a classical Achilles' heel, often referred to as the "viable rim". The tumor periphery being mainly supplied by the normal host vasculature undestroyed by the VDA, helps sustain nearby tumor cells. Thus VDAs alone tend to be non curative as a single therapeutic agent [3, 15] Hence, combination of VDAs with other therapies is a logical step in order to achieve complete tumor therapy. The rationale behind combination of a VDA such as CA4P with radiation is well explained by Landuyt et al  as follows. They hypothesized an ad ditive antitumor effect with the combination of the direct tumor cell kill achieved by irradiation and the majorly indirect tumor necrosis due to CA4P. Their logic being that the ionizing radiation would be more effective, especially in case of large tumor s, in the proliferating and host vessel supplied tumor periphery region, whereas CA4P would achieve indirect tumor cell death, including that of the hypoxic radio resistant cells. Corroborating similar additive co operation, Li et al
21 have reported that CA4 P was able to significantly improve tumor therapy when administered 0.5 1 hour post radiation or chemotherapy  An giogenesis inhibiting agents are unlikely be used as a single agent tumor therapeutic s either. Tumor progression during the course of trea tment is a major reason for therapy failure. Tumor growth and metastasis being dependent on angiogenesis, the merit of AIs mostly lies in disease control as opposed to tumor cell kill [23, 55] Evidence suggests that cytotoxic therapies including radiation therapy can initiate tumor angiogenesis  Hence the need for combination of AIs wi th conventional treatment is justified. Several studies so far have demonstrated that scheduling of the VTAs with traditional chemo and radio therapy produce better outcomes than using VTAs alone [25, 57 59] Furthermore, the combination of AIs with VDAs is believed to have potential for even more compreh ensive tumor therapy [24 26] Since the A Is, VDAs and conventional treatments such as chemo and radio therapy accomplish tumor damage via altogether different mechanisms and can have non interfering treatment schedules; it makes sense to utilize these means to treat the tumors by attacking from as many fronts as possible. Even though, at first thought, it would seem that VDAs and AIs might result in hindered tumor blood perfusion and reduced oxygen concentration by disrupting tumor vasculature or reducing tumor vessel density, several studies ha ve shown conflicting data regarding tumor perfusion and oxygenation as well summarized in the review by Horsman et al.  The effects of VDAs on tumor pathophysiology are relatively less controversial with typical reduction of tumor perfusion between 1 6 hours post VDA and
22 recovery at later time points, with the degree and duration of the vascular shut down being dependent on the drug type, dosage, and tumor type  On the contrary, AIs have been rep orted to induce extremely contradicting results in terms of vessel density, tumor oxygenation, and blood perfusion in different tumors evaluated using different methods  Therefore we concluded that f urther investigation and in depth ch aracterization of VTA response wa s needed for their better utilization in tumor therapy. Evaluation Tools for Vascul ar Targeting Therapy Tumor oxygenation measurement is considered as a classical evaluation endpoint from techniques such as the Eppendorf polarographic oxygen electrodes [60 62] hypoxic markers [63 65] or radiation response assays [66 68] for characterization of AI efficacy, often times with conflicting results. On the other hand characterization of VDA efficacy is performed in preclinical and clinical studies, by using convention al assays for tumor perfusion, blood flow, vascular structure, vessel permeability, and vascular damage in tumor and normal tissue [28, 29, 69 71] Though useful, some of these methods are unable to provide real time serial information in vivo, and require animal sacrifice at the point of measurement. For observing real time in vivo changes after VDA treatments, Sheng et al used mi crosphere fluorescence and assessed tumor blood flow in murine flank tumors after OXi4503 treatment  Several other techniques such as MRI and PET can also be used to observe bloo d flow changes, tumor perfusion and vascular permeability [3, 72] but are unable to provide microvascular resolution. To obtain microvessel function data and to observe serial morphological changes after VDA treatment, Tozer et al have employed multi photon fluorescence microscopy measurements in their window chamber models 
23 We t hought it important to look at microvessel oxygenation for VTA characterization, not only because they induce oxygenation fluctuations, but also due to the relevance of tum or oxygen status for other therapeutic modalities. As we know, oxygenation of a vessel depends on several factors such as hemoglobin saturation (HbSat), perfusion, and red cell flux [74, 75] Spectral imaging enabled us to provide direct information regarding microvessel oxygenation in terms of HbSat. Motivation and Goal of Research Develop ment of novel VTAs holds promise for effective tumor treatment. Combination of VTAs with conventional therapies or with each other, have been shown to have additive or supra additive effects on tumor control and therapy Thorough preclinical evaluation of these drugs in live animal models is essential prior to human trials. Pathophysiological changes post VTA treatment in terms of structural and vessel function changes are important parameters to characterize the treatment efficacy and optimize treatment re gimen Despite the abundance of information regarding these parameters acquired using various techniques, there remains a need for a quantitative, real time, and direct observation of these phenomenon in live animals. Through this research we aspired to de velop a spectral imaging based mouse tumor system for real time in vivo microvessel structure and functional measurements for VTA characterization. A model tumor system for window chamber studies was identified, and then combinatorial effects of VDA and AI were characterized in model tumor system. Specifically, we studied the effect of individual as well as combinatorial treatment of human renal carcinomas (Caki 1 tumors) with two different VTAs (OXi4503, a tubular binding vascular disrupting agent; and Ava stin an anti VEGF angiogenesis inhibitor) using our mouse window chamber model and spectral imaging system. We were able to
24 provide further insights in the working of these drugs and study their effects on tumor vessel pathophysiology. The eventual goal o f this research is the in depth characterization and optimal combinatorial scheduling of VTAs with conventional therapies.
25 CHAPTER 2 MOUSE MODEL AND SPEC TRAL IMAGING SYSTEM : AN INTRODUCTION Window Chamber Mouse Model : Background The rodent dorsal skinfol d window chambers have been used for a long time to non invasively study blood vessels. The chamber model has been used for microcirculatory studies [76, 77] study of endometriosis  and studies of tumor microvasculature and angiogenesis [79 81] Early tumor growth and angiogenesis has been recorded by Li et al  by injecting 20 to 30 fluorescently labeled tumor cells into the window chamber and monitoring the tumor growth right from the initiation stages, along with tumor cell migration toward and around the existing host microvessels. Modified window chambers have been successfully used in rats, hamsters, as we ll as mice [76, 77] I maging techniques such as confocal and multiphoton microscopy have been used with the window chamber to image fluorescent ly labeled cells as well as unlabeled extracellular matrix components [83 85] We used t he window chamber model in our work to look at tumor s, tumor and normal microvasculature, blood hemoglobin saturation and blood flow. Several intra vital imaging modalities such as brightfield imaging, fluorescent imaging, Laser scanning microscopy and s pectral imaging were successfully employed in live tumor bearing mice. Our window chamber mouse model with t he window chamber implanted on a nude mouse is depicte d in Figure 2 1.
26 Figure 2 1. The window chamber model. Figure 2 1a shows the window chamber implanted on a healthy nude mouse. Figure 2 1b demonstrates the 4T1 tumor growing at the implantation site in the window chamber indicated by the dashed circle, several days after implantation. b a Tumor 1cm
27 Spectral Imaging: Background Spectral imaging is a very advantageous combination of spectroscopy methods with imaging modalities. It is a great tool that allows us to simultaneously study multiple features such as organelles and proteins both qualitatively as well as quantitatively. E ven though b oth spectroscopy and imaging te chniques are well established in their respective fields, their combination is a relatively novel concept with great possibilities yet to be explored The need to measure the spectrum each point of the image often leads to long acquisition times. For practical use in biomedical applications, compromises are made to reduce the acquisition times in a tolerable range  Spectral imaging requires imaging the same object multiple times, with each image being taken at a different wavelength, thus creating a three dimensional data set. As depicted in Figure 2 2 the spectral data acquired at each pixel is the light intensity at each wavelength, I(x,y, ). It can be viewed either as an image I(x,y) at each wavelength k, or as a spectrum I(k) at every pixel (x,y).
28 Figure 2 2 Description of a spectral image data set  As described by Garini et al. each point in the 3 D spectral image cube represents a single number and the spectral image is described as I(x,y,k). It can be viewed either as an image I(x,y) at each wavelength k, or as a spectru m I(k) at every pixel (x,y). The image acquisition time highly depends on the detection device, which could be a single point detector, a line detector, or a 2 D array detector such as a charged couple device (CCD). Obviously, if any of these methods are used for image detection, the spectral image acquisiti on could not be performed at once. There are several different strategies employed for spectral imag scan an d
29 For obtaining spatial scans, the entire spectrum of a portion of the image at a time is measured and the entire image is scanned in such manner. Time scan methods measure a set of images where individual images are a superposi tion of spectral or spatial image information. The actual spectral information is then transformed in the actual spectral image using methods such as Fourier transforms. Moreover, the whole spectrum measurement methods involve the measurement of entire spe ctral image at once, but in order to do so the spectral or spatial resolution is compromised. In wavelength scan methods, one image at one wavelength is captured at a time. For a small number of wavelengths color filters can be used. Otherwise, any of the variable filters such as the circular variable filter (CVF)  liquid crystal tunable filter (LCTF)  and acousto optical tunable filter (AOTF)  may be used. These filters can capture a full spectral image by measuring one image at a time but each time at a different wavelength. They enable us to acquire a user selectable number of wavelength images with adjustable exp osure time for each separate wavelength. Imaging Acquisition and Image Processing: Techniques Used in the Presented Research Imaging System A Zeiss microscope (Carl Zeiss, Inc., Thornwood, NY) was used as the imaging platform (Figure 2 3 ) For transillumi nation of window chambers, a 100W tungsten halogen lamp was used. Images were obtained at 1380 x 1035 pixels and 12 bit dynamic range using a CCD camera thermoelectrically cooled to 20C (DVC Company, Austin Texas; Model # 1412AM T2 FW). The long working distance objectives used were 2.5x and 5x fluars, 10x EC Plan NeoFluar, and a 20x LD Plan NeoFluar (Carl Zeiss, Inc., Thornwood, NY). Hyperspectral images were obtained via band limited
30 optical filtering using a C mounted liquid crystal tunable filter (LCT F) (CRI, Inc., Woburn, MA) with a 400 720nm transmission range and a 10 nm nominal bandwidth, placed in front of the camera. Images were saved as 16 bit TIF files. Fluorescent images for 4T1 GFP were taken using a FITC filter set (Carl Zeiss, Inc., Thornwo od, NY; Excitation range: 450 490nm; Emission range: 515 565nm) at 520nm. Figure 2 3 The Imaging system. The Zeiss based imaging microscope system is shown in Figure2 2a. Figure 2 2b gives a close view of the mouse being im aged, placed on the heating pad and breathing through the nose cone. Image Acquisitio n LabVIEW8 (National Instruments Corp., Austin, TX) was used to prepare a custom designed virtual instrument for controlling the tuning of the LCTF filter and operation o f the CCD camera. The software enabled automated image acquisition using the a b
31 specifications for camera exposure time and gain for each filter wavelength. Since the LCTF filter transmits less at lower wavelengths and more at higher wavelengths the exposure time for the camera had to be controlled such that the full dynamic range of the camera was utilized. The minimum exposure time used was 400ms whereas the maximum exposure time used was 1400ms, resulting in a typical acquisition time of approximately 16ms for image acquisition, filter tuning, image transfer, and saving images on external hard drive. One hemoglobin saturation image set comprised of 16 images acquired in the wavelength range of 500 575nm with an interval of 5nm. Image Processing Image proces sing was performed using Matlab software (The Mathworks, Inc., Natick, MA). The hyperspectral images acquired using the CCD camera and LCTF filter were converted into double precision arrays for mathematical analysis. The mathematical analysis was based on saturation (HbSat) using linear least squares regression  The model equation used is the following: Where, OD = optical density (absorbance) at wavelength i (no units), eHbO2 = extinction coefficient of oxyhemo globin at wavelength i (1/(M.cm)), [HbO2] = concentration of oxyhemoglobin (M) which is also the hemoglobin saturation in this form of the equation, eHbR = extinction coefficient of deoxyhemoglobin at wavelength i (1/(M.cm)), L = pathlength (cm),
32 LS = s cattering term (dimensionless here since it is the product of the pathlength and the scattering). In this equation, the scattering is handled in the LS term. A rearrangement of the OD function results in the following: Where, HbSat = hemoglobin saturation (= [HbO2]/ ([HbO2] + [HbR])), and HbTotal = [HbO2] + [HbR]. If the following substitutions are made, a = HbSat*L*HbTotal, b = L*HbTotal, c = LS, eDelta(i) = eHbO2(i) eHbR(i), then the following equation results: The function solves this equation for a, b, and c by linear least squares regression, and then solves for HbSat by performing a/b. The image processing flow di agram is presented in Figure 2 4
33 Figure 2 4 Flow diagram depicting the image processing method to obtain hemoglobin saturation map from the stack of 16 images obtained via spectral imaging. Further details regarding the imaging system and image processing can be obtained from Sorg et al  Hemoglobin Saturation map linear least squares regression model HbO 2 and HbR ratio determined by the ligh t intensity 5nm intervals 500 575nm 16 images
34 CHAPTER 3 SPECTRAL IMAGING OF WINDOW CHAMBERS: A VERSATILE TOOL FOR VASCULAR PATHOLOGY S TUDIES In the previous chapter an introduction to our mouse window chamber model and spectral imaging system was provided. This chapter will now furnish some of our data illustrating the versatility and utility of spectral imaging for study of microvasculature. Abnorma l vasculature aberrant angiogenesis and oxygenation, or abnormal vascular remodeling are characteristic features of numerous patho logies. Qualitative and quantitative imaging and measurement of microvessel function may promote increased understanding of these diseases. Spectral imaging techniques can be effectively used for direct imaging of microvascular function. We used spectral imaging to study microvasculature physiology and function in live mouse models of various pathological conditions such as cancer tumors thrombosis and arterio venous malformations. Introduction Imaging and measurement of microvessel function can be imp ortant to increase understanding of various diseases. Numerous pathologies include hypervascularity, aberrant angiogenesis, or abnormal vascular remodeling among the characteristic features of the disease [91 93] For example, the microvasculature of solid tumors is structurally and functionally abnormal [94, 95] foot ulcerations can occur in diabetes mellitus patients due to impaired angiogenesis and poor vascularizat ion [96, 97] angiogenesis that forms abnormal microvasculature can sustain chronic inflammation in rheumatoid arthritis [98, 99] and epidermal hyperplasia is a characteris tic of psoriatic skin lesions  Many ophthalmic diseases, such as diabetic retinopathy and sensile vascular degeneration, also have an underlying microvascular component [101, 102]
35 Several optical techniques are useful for direct imaging of microvascular morphology and function either clinically or in preclinical animal models. Intravascular fluore scence contrast agents can be used for imaging of microvessel morphology in some tissues with wide field fluorescence imaging (e.g., retinal angiography with sodium fluorescein  ) or multiphoton microscopy using dyes or quantum dots  Multiphoton microscopy can also be used to measure microvessel blood velocity  permeability  and relative oxygenation  in animal models. Relative microvessel pro fusion can be measured with laser Doppler and speckle contrast techniques [108, 109] Photoacoustic tomography and microscopy can be used to measur e microvascular morphology and relative vessel oxygenation [110, 111]  Optical coherence tomography (OCT) can be a par ticularly versatile optical modality for microvasculature imaging and measurements. Blood velocity can be measured [113 115] with Doppler OCT, relative microvessel oxygenation can be measured with spectroscopic OCT methods [116 119] microvessel hematocrit can be measured from scattering attenuation by red blood cells  and speckle variance processing can provide contrast for microvessel morphology imaging  Spectral imaging can be used to assess microvascular oxygen transport function in microvessel networks through measurements of hemoglobin saturation. Spectral imaging of microvessels with high resolution is limited to superficial tissue microvessels and the best results are often obtained invasively with animal models; thus, the utility of spectral imaging measurements of micro vascular function may potentially be applied best in the preclinical setting. Depending on instrument capabilities and the animal model used, correlations in hemoglobin saturation can be obtained throughout a
36 microvessel network with high spatial and tempo ral resolution. Several researchers have used spectral imaging in preclinical models to measure microvascular hemoglobin saturation in mesentery tissue  cremaster muscle [12 3] rodent dorsal skin fold window chambers [122 126] and cerebrum [109, 127] In this chapter, we highlight novel observations made with our intravital microscopy spectral imaging system used with mo use dorsal skin fold window chambers for imaging hemoglobin saturation in microvessel networks. Specifically, we image acute oxygenation fluctuations in a tumor microvessel network the formation of spontaneous and induced microvascular thrombosis and occl usions. Materials and Methods Details about the mouse model and spectral imaging system can be found in previous chapter Methods for Tumor Vessel Acute Oxygenation Fluctuations Preparation of fluorescent red blood c ells RBCs were fluorescently labeled by using a modification of the procedure by Unthank et al.  RBCs obtained from donor mice wer e labeled with a 1mg/ml stoc k dioctadecyl tetramethylindodicarbocyanine, 4 chlorobenzenesulfonate salt (DiD solid, Invitrogen, D 7757) dissolved in ethanol. DiD (excitation 644nm=emission 665nm) was chosen as the lipid membrane labeling sol ution because it is an analog of the commonly used DiI dye (ex 549nm, em 565nm), but with a markedly redshifted fluorescence excitation and emission spectrum to minimize interference of RBC flux measurements with hemoglobin saturation measurements (500 575 nm). RBCs obtained by cardiac puncture from a donor mouse were washed twice via centrifugation and resuspension in phosphate buffered saline
37 to 10 ml of sterile PBS. T he solution was incubated at room temperature for 30 min and agitated every 10 min to ensure suspension of the RBCs. After labeling, the RBCs were washed to remove unbound dye and re suspended in PBS. Immediately prior to an of packed labeled cells in saline solution (30% V/V) was administered to the mouse to be imaged by tail vein injection. An aliquot of DiD labeled cells was saved for flow cytometric confirmation of the efficacy of the labeling procedure. Preparation of 4TO 7 t umor c ells We used 4TO7 mouse mammary adenocarcinoma cells that are non metastatic subclones of the 4T1 cell line with athymic female nude (nu/nu) mice. The tumor cells were a gift from Mark W. Dewhirst, Duke University, Durham, North Carolina. The tumo r cells were cultured as a monolayer in Dulbecco modified Eagle medium (DMEM; Mediatech, Manassas, Virginia) with 10% fetal bovine serum (Mediatech, Manassas, Virginia) prior to implantation. Cultures were used after one or two passages from frozen stocks to ensure recovery from the thermal shock and a normal growth rate. The cells were enzymatically dissociated from the flasks (BD Bioscience, San Jose, California) using 0.05% trypsin/EDTA (Mediatech, Manassas, Virginia) and counted on a hemacytometer to de termine the cell concentration to prepare single cell suspensions for implantation. The 4TO7 tumors were established at the time of window chamber surgery from a 10 L single cell suspension of 5 10 3 to 10 10 3 cells injected into the subcutaneous tissue imm ediately prior to placing a 12 mm round glass coverslip over the exposed area of the skin.
38 Window surgery, in vivo t umor i nitiation and t umor i maging A titanium window chamber was surgically implanted under anesthesia (ketamine 100 mg/kg IP (intraperitonea l) and xylazine 10 mg/kg IP) on the back of mice. In this window chamber model, one piece of skin was removed completely and replaced with a 12 mm diam #2 round glass coverslip (Erie Scientific, Portsmouth, New Hampshire). For experiments with tumors, a wi ndow chamber tumor was established during chamber implantation by injecting 10 L of a single cell suspension of tumor cells. In survival experiments, animals were housed in an environmental chamber with free access to food and water and standard 12 h ligh t/dark cycles. For imaging, animals were placed on a heating pad attached to the microscope stage during the imaging session. A custom built window chamber holder stably secured the window chamber under the microscope objective. The holder was fastened to the microscope stage in a manner that enabled the window chamber to be positioned under the microscope objective using the standard manual controls on the microscope stage. Anesthesia for immobilization during ima ging was provided by isoflurane (1 to 1.5%) in air. In two experiments, an increase in convective oxygen transport was induced during imaging by a change in breathing gases from room air to 100% oxygen. Image and data p rocessing Image processing was performed using software developed with MATLAB (T he Mathworks, Inc., Natick, Massachusetts ). Images were converted into double precision arrays for mathematical processing. Raw pixel values were converted to absorbance values after manually selecting a vascular reference region in the images as estimate s of unattenuated light.
39 Calibration spectra of oxy and deoxyhemoglobin obtained with the imaging system were used to calculate pixel HbSat values by linear least squares regression of the data in a linear mixing model according to the method of Shonat e t al.  as described prev iously  Red blood cell f lux i maging Fluorescently labeled red blood cells (RBCs) were imaged via streaming video using an Andor iXon electron multiplying CCD camera (Andor Technology, South Windsor, Connecticut). A Cy5 filter set was used (Chroma T echnology Corp., Rockingham, Vermont, excitation 640nm with 20nm bandwidth, emission 680nm with 30nm bandwidth) in line with the illumination source of a Zeiss FluoArc mercury lamp. To optimize the resolution of the fast moving cells, data were acquired in kinetic acquisition mode wit h 2 2 binning, using an exposure time of 16:2ms, a shift speed of 30 Hz to ensure that images would be captured with sufficient temporal resolution. Internal triggering was used to spool data directly to the computer hard dr ive, and 20 s of streaming video was saved every minute for 1 h in coordination with the acquisition of spectral imaging data sets. Each frame was saved as a tagged image format file (TIFF), resulting in a stack of 600 TIFFs per data point. RBC flux was de termined by first identifying a region on a vessel of interest where flowing RBCs appeared to be in good focus and then manually counting the number of labeled cells flowing past the designated location on the vessel over the 20 s time interval of data acq uisition for the time point. These measurements were converted to flux by correcting for the fraction of labeled RBCs versus unlabeled RBCs in the mouse blood. The fraction of labeled RBCs was determined by flow cytometry of a blood
40 sample obtained postima ging via retro orbital puncture performed on the imaged mouse. The RBC flux was then calculated by using Eq. (4): ...................(4) where, N RBC is the number of fluorescent RBCs counted flowing in the vessel, t is the time in terval in which N RBC was counted (20 s in this case), and N RBC fluor and N RBC,T otal are the numbe r of fluorescent and total RBCs respectively, counted by flow cytometry in the post experiment blood sample. RBC velocity was determined from the time required for an individual RBC to travel a specific distance in the blood vessel that passed through the region of interest. The time interval was determined from the frame rate of fluorescence video imaging. The mean velocity of 5 10 randomly selected RBCs from e ach time point was taken as vavg. Methods for Induced Thrombus Experiments Experiment d esign s ummary The FeCl3 was applied to the tissue with a piece of paper soaked with the solution (dotted outline in Figure 4 1). Venous thrombi gradually formed over app roximately 20 min, resulting in a decreased oxygenation in the venules due to restricted and occluded blood flow. Figure 4 2 shows time series images of the region outlined by the small box in Figure 4 1, showing the gradual decrease in oxygenation of the venule as thrombi form over time. The HbSat values of the indicated region in Figure 4 2 at several time points are shown in Figure 4 3. Thrombus formation was independently confirmed by
41 the accumulation of fluorescently labeled platelets throughout the ve nules. Spectral imaging revealed the microvessel network oxygenation changes due to thrombi formation in the venules. Imaging of thrombus formation was performed as part of experiments to test the antithrombotic activity of an angiotensin converting enzyme 2 (ACE2) activator compounds to stimulate purported antithrombotic activity of the ACE2/angiotensin (1 7)/Mas axis [130, 131] Vascular injury and thrombus formation were artificially induced to demonstrate the effect of the compounds by visualizing platelet adhesion and thrombus formation in real time. Surgery Dorsal skin fold window chambers were surgically impla nted on male athymic nude mice ( nu/nu ) of 9 to 10 weeks of age ( Harlan Sprague Dawley, Indianapolis, Indiana ) as previously described except that a coverslip was not installed for these terminal experiments. The exposed subdermal skin of the window chamber wa s maintained under warm saline at 37 C throughout the experiment. Platelet labeling and visualization The jugular vein was cannulated for intravenous access. A bolus of 300 L of carboxyfluorescein diacetate succinimidyl ester (Invitrogen, Carlsbad, Ca lifornia) was injected via the cannulated jugular vein to visualize platelets. The nonfluorescent precursor is taken up by platelets and somewhat by leukocytes where it forms the stable and highly fluorescent fluorophore carboxyfluorescein succinimidyl est er (CFSE) in the cells by intracellular cleavage of the acetate groups. The approximate excitation and emission peaks of CFSE are 492 and 517 nm, respectively.
42 Thrombus i nduction Thrombus formation was induced using a protocol similar to Katayama et al.  via topical application for 2 min of 5% FeCl 3 soaked paper (3 3 mm) placed on the exposed sub dermal skin region containing the vessel of interest. The piece of paper was removed after 2 min and the window was wa shed twice with warm saline. Thrombus i maging A vessel of interest with a 70 to 200 m diameter was selected on the basis of un restricted blood flow and location in the vessel network. Thrombus formation was monitored in real time by fluorescence and s pectral imaging. Fluorescence video images were acquired every minute for 20 s at a 30 frames/s rate with an electronmultiplying CCD camera (Andor Technology, South Windsor, Connecticut) and a spectral imaging data set for hemoglobin saturation measureme nts was acquired once every minute, as described previously. Imaging sessions lasted for 20 to 25 min. Methods for A rterio venous Malformation Experi ments Experiment design s ummary Imaging of pathological anastomoses formation in the form of arteriovenous malformations was performed using a mouse model of hereditary hemorrhagic telangiectasia (HHT). In this mouse model, the activin receptor like kinase 1 (Alk1) gene is designed to be deleted by administration of tamoxifen. Details about the generation of th e Alk1 conditional knockout mice were described previously  Conditional Alk1 knockout (Alk1 ) mice were intercrossed with ROSA26 CreER/+ mice that ubiquitously produce CreER that is a fusion protein of Cre recombinase and a mutated form of the estrogen receptor that can be activated by tamoxifen. The experimental group included
43 mice with ROSA2 6 CreER/+ ; /Alk1 2f/2f and ROSA26 CreER/+ ;Alk1 genotypes while the control groups were ROSA26 +/+ ;Alk1 2f/2f and ROSA26 +/+ ;Alk1 mice. Window chamber wound m odel Window chambers were surgically installed as already described, and a wound was created in an avascular area between vessel branches in the center of the window chamber using a 16 gauge needle during surgery. Oil based tamoxifen (Sigma Aldrich, St. Louis, Missouri) was administered intraperitoneally at 2.5 mg/25 g bodyweight to control and kn ockout mice immediately prior to window chamber installation. Spectral i maging Mice were imaged as already described almost every day for up to 11 days after surgery. Blood flow imaging Blood flow through suspected arteriovenous malformations (AVMs) from a rterioles to venules was confirmed by tracking fluorescently labeled red blood cells (RBCs). Blood from a donor mouse was labeled with the carbocya nine dye 1,1 dioctadecyl 3,3,3,3 tetramethylindodicarbocyanine, 4 chlorobenzenesulfonate salt (DiD solid, Inv itrogen, D 7757) similar to the procedure used by Unthank et al  A 50 L bolus of packed lab eled RBCs in saline solution (30% v/v) was administered by tail vein injection 1 to 2 days prior to imaging. It was determined from previous experiments that for the fraction of labeled cells achieved in vivo with this protocol (1 to 3%), there was negligi ble absorption interference of the dye with the spectral data obtained for hemoglobin saturation in the wavelength range employed 40 (500 to 575 nm).
44 Results Tumor Microvascular Acute Oxygenation Fluctuations Abnormal tumor microvasculature with compromise d oxygen transport is known to be a major cause of hypoxia in tumors [31, 95, 134] The biology of tumor cells may be different depending on whether they are exposed to chronic hypoxia or acute hypoxia with subsequent reoxygenation  Acute fluctuations in blood flow and oxygen supply that are significant enough to cause hypoxia are becoming more appreciated for their effects on tumor biology and therapy as even relatively short hypoxic episodes can have a potentially profound effect [136, 137] For example, it has been recently reported that tumor metastases have a higher fraction of acutely hypoxic cells than chronically hypoxic cells, implying that acutely hypoxic cells have a higher metastatic potential and greater probability of impact on therapeutic response  Research to characterize acute fluctuations in tumor microvessel oxygenation and perfusion requires advanced imaging techniques with microvessel resolution  Spectral imaging of tumor microvessel oxygenation may be a useful tool in this effort. We previously showed that spectral imaging could be used to measure acute fluctuati ons in tumor microvessel oxygenation  We further demonstrate here how spect ral imaging can document acute fluctuations in tumor microvessel oxygenation and reveal an apparent sensitivity in a tumor microvessel network to supply vessel oxygenation fluctuations. Figure 3 1 shows a 4TO7 tumor with a diameter of 1.5 to 2.0 mm 9 days after implantation of cells. Temporal spectral data was collected every minute for 1 hour. Figure 3 2 shows the HbSat image created from one set of spect ral data from the data in Figure 3 1. Also, red cell flux data was obtained simultaneously every minute for 1
45 hour. An example figure is shown in Figure 3 3. The regions of interest (ROIs) chosen for HbSat measurement are indicated in the figures by number ed white squares in Figure 3 1 RBC flux was calculated after counting the number of labeled RBCs flowing in the vessels close to the ROIs of interest. The plot of HbSat values and RBC flux versus time is shown in Figure 3 4 and Figure 3 5. Correlation bet ween the RBC flux and HbSat c an be observed These results demonstrate that fluctuations in microvessel oxygenation with the potential to cause hypoxia can occur in tumors of a relatively small size. The results also suggest that a potential source of inst ability in tumor microvessel oxygenation may be due to an exaggerated response in the tumor vessels to oxygenation fluctuations in the tumor supply vessels. Gaustad et al. showed that tumor feeding vessels can have a significant influence on local perfusio n of tumor microvessel network regions  thus, some correlation between tumor supply vessels and regions of tumor microvessel n etworks could be expected. In previous publications, we showed that in tumors there could be direct connections between well and poorly oxygenated microvessels [140, 142] If these connections were present in the tumors used in this study, then it is possible that dynamic changes in blood flow through these connections contributed to the large swings in microvessel oxygenation.
46 Figure 3 1 Transmitted light image of the tumor microvessel network. It is a 4TO7 mouse mammary carcinoma with a diameter of 1.5 to 2.0 mm 9 days after implanta tion of cells. White triangles represent regions of interest chosen for analysis. A 10 objective was used for imaging (NA of 0.3, working distance of 5.5 mm).
47 Figure 3 2. Hemoglobin saturation image of the tumor microvessel network in Figure 3 1. Th e pixels are colored according to the hemoglobin saturation scale to the right of the figure. The background is black. Regions of interest for analysis are indicated in the figure.
48 Figure 3 3 Fluorescence image of RBCs labeled with DiD flowing in the tumor microvessel network. Images were captured with an electron multiplying CCD camera at video rates (30 Hz). 20 s of data were taken every minute of the hour long imaging session, and cells were counted over this time increment and then converted to RB C flux measurements. The locations of the regions of interest for analysis are indicated in the figure.
49 Figure 3 4. Plot of RBC flux versus hemoglobin saturation (HbSat) for region of interest 1 in Figure 3 1. The data points were acquired at 1 m in intervals for 1 h. Figure 3 5. Plot of RBC flux versus hemoglobin saturation (HbSat) for region of interest 3 in Figure 3 1. The data points were acquired at 1 min intervals for 1 h.
50 Spectral Imaging for Study of Microvascular Occlusions Blood vesse l occlusions due to thrombus formation can occur in disease and pathophysiological processes. In cardiovascular disease, the eruption of an atherosclerotic plaque can initiate arterial thrombus formation leading to myocardial infarction  There can be a risk of deep vein thrombosis formation a fter some types of surgery or trauma, or in patients with various disorders of the coagulation system  Imaging of microvasculature in preclinical models can potentially be useful for testing and evaluation of new thrombolytic drugs to dissolve thrombi or thromboprophylaxis agents aimed at preventing thrombus formation. Dorsal skin f old window chambers were used with spectral imaging as part of an investigation into the antithrombotic activity of a novel ACE2 activating agent [145, 146] Dorsal skin fold window chambers were used with spectral imaging as part of an investigation into the antithrombotic activity of a novel ACE2 activating agent. Figure 3 6 shows brightfield and HbSat images of a re gion of tissue in a mouse immediately prior to and 21 min after thrombus formation initiated by topical application of FeCl3. The FeCl3 was applied to the tissue with a piece of paper soaked with the solution (dotted outline in Figure 3 6 ). Venous thrombi gradually formed over approximately 20 min, resulting in a decreased oxygenation in the venules due to restricted and occluded blood flow. Figure 3 7 shows time series images of the region outlined by the small box in Figure 3 6 showing the gradual decrea se in oxygenation of the venule as thrombi form over time. The HbSat values of the indicated region in Figure 3 7 at several time points are shown in Figure 3 8 Thrombus formation was independently confirmed by the accumulation of fluorescently labeled pl atelets throughout the venules. Spectral
51 imaging revealed the microvessel network oxygenation changes due to thrombi formation in the venules. Figure 3 6. Brightfield and HbSat images are shown before and 21 minutes after topical application of FeCl3 t o induce thrombus formation. The red dotted line in the top left image indicates where the border of the paper soaked with FeCl3 was relative to the image area. The yellow box indicates the region shown in the images in Figure 3 7 at higher magnification.
52 Figure 3 7. Brightfield and HbSat time sequence images of thrombus development for the region indicated by the yellow box in Figure 3 6 The yellow outlined area on the venule in the top left image indicates the ROI used for the HbSat measurements show n in Figure 3 8 The arrow in the brightfield image at the 8 min time point indicates a forming thrombus.
53 Figure 3 8. Plot of the HbSat in the ROI indicated in Figure 4 2 at various time points during thrombus development. The data points in the figure are given as the medianinterquartile range for the ROI. Spectral Imaging for Real Time Imaging of Arterio Venous Malformations Arteriovenous (AV) anastomoses are direct connections between the arterial and venous circulation without an intervening capil lary bed. These connections can be created surgically in larger vessels for a therapeutic purpose such as assistance in the treatment of hemodialysis patients  AV anastomoses are naturally present in the microcirculation of some normal tissues with highly variable blood flow, such as skin  and nasal mucos a  but they can also arise pathologically as part of a disease process. Tumor microvasculature can develop a substanti al number of pathological AV anastomoses  A high percentage of HHT patients can develop AVMs in the brain, lungs, liver, and gastrointestinal tract that can potentially rupture with devastating consequences [175,176] In addition to these visceral AVM s, a majority of
54 HHT patients develop telangiectases in mucocutaneous areas, such as the nasal cavity, that result in spontaneous nose bleeding  To study the development of skin AVMs, we developed a mouse model using the Alk1 gene by intercrossing RO SA26 CreER/+ mice with Alk1 conditional knockout (Alk1 2f ) mice  We have shown that a wound can induce de novo AVMs in its surroundings in adult ROSA26 CreER/ + ;Alk1 mice (Alk1 mutants hereafter) administered with tamoxifen. To study the development of skin AVMs, we developed a mouse model using the Alk1 gene by intercrossing ROSA26 CreER/+ mice with Alk1 conditional knockout (Alk1 ) mice  We have shown that a wou nd can induce de novo AVMs in its surroundings in adult ROSA26 CreER/+ ;Alk1 mice (Alk1 mutants hereafter) administered with tamoxifen. Figure 3 9 shows AVM development at selected time points stimulated by wound healing in the Alk1 mutants. The region s where two AVMs form are indicated in the figure. Venules near the wound had increased oxygenation and diameters after surgery due to AVM formation. Veins and venules, the most distensible of the blood vessels, are sensitive to pressure changes and can ma rkedly dilate with small increases in pressure  The increase in venule oxygenation was observed prior to the appearance of a clearly identifiable AVM with brightfield or spectral imaging. Confirmation of a direct connection between arterioles and venules through AVMs was obtained by tracking the flow of fluorescently labeled RBCs administered via tail vein injection after spectral imaging. Tracking fluorescently labeled RB Cs also enabled visualization of the earliest connections between arterioles and venules through sprouting vessels that eventually formed AVMs, suggesting that the vascular changes
55 seen in the brightfield and spectral images were induced by these connectio ns. In the Alk1 mutants, relatively few nascent sprouts developed at the perimeter of the wound. There was extensive formation of stable AVMs that resulted in significant changes in the microvasculature adjacent to the wound, including major arterioles and venules, that persisted after the wound had healed. Wound healing occurred within the same time frame as control animals. Figure 3 10 shows wound healing at selected time points in control animals with functional Alk 1. Extensive formation of nascent spro uts occurred around the periphery of wounds in control animals during wound healing, but they subsided after the wound was healed. There were minor changes in the oxygenation and morphology of major vessels and after wound healing the microvasculature retu rned to a more normal state. In contrast to the Alk 1 deletion mice, no AVMs formed in control mice and venule oxygenation in control mice was less than in Alk 1 deletion animals. This point is illustrated in Table 3 1. Arteriole and venule ROIs were evalu ated on different days (n=9 for Alk 1 deletion, n=7 for control). The ROIs were located in areas near the arrow heads identifying the different vessels in Figs. 3 9 and 3 10 For the mice in the figures, the arterioles had similar average saturations throu ghout the observation period but the Alk 1 deletion mice had statistically higher venule saturation (see Table 3 1).
56 Fig ure 3 9. AVM development in a conditional Alk1 deletion mouse ( ROSA26 CreER /Alk1 ) stimulated by a wound, where A, arteriole; V, venule; and W, wound. The left column shows transmitted light images and the right column shows HbSat images. Region where AVMs formed is indicated in the day 5 and day 9 images. At day 5, the AVM connect ions were not obvious but clear changes in venule oxygenation could be seen due to the AVMs. The arrowheads indicating the arteriole and venule also indicate regions of interest where HbSat measurements were obtained.
57 Figure 3 10 Wound healing in a co ntrol mouse: A, arteriole; V, venule; and W, wound. The left column shows transmitted light brightfield images and the right column shows HbSat images. The arrowheads indicating the arteriole and venule also indicate regions of interest where HbSat measure ments were obtained. Unlike the conditional Alk1 deletion mouse in Figure 3 9 no AVMs formed during wound healing.
58 Table 3 1 Comparison of arteriole and venule saturation in a mouse model of HHT during wound healing. Discussion Several optical tech niques are useful for direct imaging of microvascular morphology and function either clinically or in preclinical animal models. Spectral imaging can be used to assess microvascular oxygen transport function in microvessel networks through measurements of hemoglobin saturation. Spectral imaging along with flurocence imaging techniques was used to study the tumor vessel oxygenation and RBC flux simultaneously as show in Figures 3 1 5. These results indicate that oxygenation fluctuations are prevalent in eve n millimeter size tumors which may potentially lead to acute hypoxic conditions in tumors. The study of RBC flux together with HbSat of the vessels gives insights regarding the relationship between the blood supply and transportation. Dangerous and debilit ating thrombus formation can be caused by pathological conditions such as diabetes, hypertension, and atherosclerosis  Thrombi induced from these and similar pathological conditions can result in strokes, pulmonary emboli,
59 and deep venous thrombi amo ng other conditions  Spectral imaging of hemoglobin saturation showed the gradual formation of chemically induced thrombi and the effects of the thrombi on microcirculation oxygenation in Figures 3 6 9 The time for thrombus formation and complete mi crovessel occlusion could be measured from these experiments, and these data may provide additional information besides systemic physiological measurements with which to assess local microvessel physiology during a thrombotic event. Real time imaging of th rombus formation and changes in microvascular oxygenation can aid in the development and comparison of new therapeutic agents designed to treat pathological thrombi by revealing subtle differences in the timing and magnitude of drug action on the microvess el physiological response and thrombus formation. Spectral imaging in a mouse model of HHT in Figures 3 9 10 and Table 3 1 document the consequences of AVM development in real time. Microvessel oxygenation changes during AVM development indicated that AVM formation occurred at early time points following angiogenesis stimulation by wound creation, and suggested that the formation of the AVM induced changes in microvascular morphology and function, particularly in the venules. Venules connected to arterioles through AVMs rapidly became dilated and hyperoxygenated compared to normal venules. The changes in venule oxygenation were frequently apparent prior to the appearance of a clearly identifiable AVM. This is important because previous histologic examination of tissue specimens suggested [179,180] that the opposite was the case that dilation of microvessels initiated or at least preceded development of AVMs. A clearer understanding of the causes and consequences of AVM formation and tracking the
60 timing of AVM formation may aid in the development of therapies to better treat diseases that exhibit this pathology. Our imaging technique may aid in understanding the causes of AVM formation and other vascular pathologies at the molecular and genetic level using gene tically altered mouse models such as the one employed in this study. A more detailed description of the animal model and characterization experiments with this model can be found in a paper by Park et al 
61 CHAPTER 4 OXI4503 T REATMENT OF 4T1 AND CAKI 1 TUMORS Introduction Tumor vasculature is essential for tumor growth, proliferation and metas tasis hence numerous methods have been developed to impair tumor vessels with the intent of cutting off the tumor blood supply and access to oxygen and nutrients. An array of specific tumor vessel targeting strategies emerged after Denekamp et al distingu ished tumor endothelial cells from normal tissue endothelial cells based on their significantly higher proliferation rate [153, 154] Currently, a number of vascular targeting approaches are in development for selective and specific targeting of the tumor vessels while sparing the normal tissue vasculature [12 16, 54] Vascular disrupting agents (VDAs) are designed to damage existing tumor blood vessels while antiangiogenic therapies such as bevacizumab aim to inhibit new blood vessel formation  Combretastatins, initially derived from the South African tree Combretum caffrum, represent a class of potent tubulin binding agents with significan t vascular disrupting activity [15, 155] Lead a gents Combretastatin A4 P (CA4 P) and OXi4503 (CA1 P) have been shown to result in time dependent decreases in tumor perfusion, increased tumor vascular permeability, and significant tumor vascular damage in a wide variety of preclinical tumor models [15, 27 29] In the clinic, these a gents have now entered Phase II/III evaluation [156, 157] Despite extensive tumor destruction following VDA therapy, a hallmark of VDA treatment is the survival of neoplastic cells at the tumor periphery  This surviving treatment [28, 158] In preclinical models and some clinical studies, functional and
62 morphological parameters such as tumor perfusion, blood flow, vascular structure, permeability and vascular damage are used to evaluate the effect of OXi4503 and CA4 P on tumor and normal tissue [28, 69, 70, 71, 158] Some of the methods used cannot furnish continuous real time in vivo information as they require animal euthanasia at the respective time points and do not permi t serial follow up of the same tumor after VDA treatment. Hence, several different techniques are used for the in vivo assessment of VDA induced functional and morphological changes in murine and human tumor vasculature. For example, Sheng et al used micro sphere fluorescence to assess in vivo tumor blood flow in murine flank tumors after OXi4503 treatment  Other techniques such as magnetic resonance imaging and positron emission t omography can provide functional information regarding blood flow, tumor perfusion, and vascular permeability [72, 158] however, with poor microvascular resolution. In vivo microscopy techniques such as multi photon fluorescence microscopy in a windo w chamber model have been employed by Tozer et al which provides vascular function data with improved microvessel resolution  and can be used to simultaneously observe serial morphological changes after VDA treatment. Spectral imaging is an effective tool for measurement of microvessel vessel oxygenation in tumor microvasculature [74, 90, 140] The oxygenation of a vessel depends on several factors such as hemoglobin saturation (HbSat), perfusion and red cell flux [74, 159] In the present study, spectral imaging is used with mouse dorsal skin fold window chambers to measure microvascular functional changes in terms of HbSat in two different tumors, a mouse mammary carcinoma (4T1) and human renal cell carcinoma (Caki 1), during serial treatments with the VDA OXi4503.
63 Materials and Methods Tumor C ells 4T1 mouse mammary adenocarcinoma cells (a gift from Mark W. Dewhirst, Duke University Medical Center, Durham, NC) were cultured in DMEM (Cellgro, Inc., 1X, 4.5 g/l glucose, L glutamine and sodium pyruvate) containing 10% fetal bovine serum (Bio whittaker, Inc.), 1% L glutamine (Clonogen, Inc.) and 1% penicillin streptomycin (Clonogen, Inc.). A single cel l suspension of approximately 5 10 3 to 10 10 3 cells prepared in serum free DMEM was used to initiate tumors in window chambers of the experimenta l mice. Caki 1 tumor xenografts  were initiated in a single hind limb of donor mice by inoculating with 2 10 5 to 5 10 5 tumor cells. Tumors were excised when they reached about 200 mm 3 in size, and a single cell suspension was prepared in DMEM. About 1 10 6 cells were used to initiate a wi ndow chamber tumor in the experimental mice. Animal M odel All in vivo procedures were conducted in accordance with a protocol approved by University of Florida Institutional Animal Care and Use Committee. Athymic (nu/nu) female mice weighing at least 21 g (Charles River Laboratories, Raleigh, NC) were surgically implanted with a titanium window chamber on the dorsal skin flap. During surgery mice were anesthetized by an IP injection of ketamine (100 mg/kg) and xylazine (10 mg/kg). A tumor was initiated in t he window chamber during surgery by injecting the tumor cells subcutaneously in the dorsal skin flap prior to placing a 12 mm diameter no. 2 round glass coverslip (Erie Scientific, Portsmouth, NH) over the exposed skin. After surgery animals were housed in 50% humidity with free access to food and water and standard 12 h light/dark cycles.
64 Drug Preparation and Experimental D esign OXi4503, a gift from OXiGENE, Inc. (Waltham, MA) was administered at a dose of 1 0 mg/kg IP (0.01 ml/g of mouse body weight). The light sensitive drug was protected from light during handling. OXi4503 was suspended in a solution of sterile saline and 50 f resh daily. Control groups received an equivalent volume of sterile saline and 5% sodium carbonate solution. The tumors were allowed to grow for 7 8 days and the treatment was started when tumors reached a size of about 3 4 mm. Spectral imaging was perform ed immediately prior to and post treatment, and also at 2, 4, 6, 8, 24 and 48 h post treatment. After the 48 h imaging time point the mouse was re treated and the same imaging schedule was maintained. The treatment was repeated 3 5 times on each of the exp erimental mice. In all, 5 mice bearing 4T1 tumors were treated with OXi4503 and 3 mice were carrier only controls. For Caki 1 tumors the OXi4503 treatment was performed on 5 mice, while 2 mice received carrier only minus drug. Imaging S ystem, Image A cquisi tion an d I mage P rocessing The spectral imaging system, image acquisition, and image processing methods for hemoglobin saturation measurements were discussed in detail previously  Briefly, a Zeiss AxioImager microscope (Carl Zeiss, Inc., Thornwood, NY) was used as a basic platform for the imaging system. For transillumination images, a 100 W tungsten halogen lamp was used. The long working distance objectives used were 2.5 and 5 Fluars, 10 EC Plan NeoFluar and 20 LD Plan NeoFluar (Carl Zeiss, Inc.). For spectral information, br ightfield transmitted light images were obtained using a CCD camera thermoelectrically cooled to T2 FW). Band limited optical filtering was accomplished using a C mounted liquid crystal
65 tunable filter (CRI, Inc., Woburn, MA) with a 400 to 720 nm transmission range and a 10 nm nominal bandwidth, placed in front of the CCD camera. LabVIEW8 (National Instruments Corp., Austin, TX) was used to automatically control the tuning of the LCTF and acquire images with the C CD camera. Microvessel hemoglobin saturation measurements and images were created from the spectral image data [89, 90] Results During each treatment the tumor vasculature experienced two distinct phases: a destruction phase closely followed by a recovery phase. The treatment resulted in a combination of three major effects on the tumor vasculature: i) transient vessel collapse, ii) permanent vessel disintegration or iii) no structural damage with only temporary functional loss. Transient vessel collapse refers to the temporary disappearance of the tumor core vessels during the destruction phase of each treatment. These original tumor vessels re emerged during the recovery phase and regained their functionality. Tumor core vasculature that underwent vessel disintegration never reappeared and in some cases there was evi dence of hemorrhaging. Some tumor vessels as well as the normal vasculature did not suffer from structural damage but still manifested temporary oxygenation changes. Treatment of 4T1 M ouse Mammary C arcinomas Each 4T1 tumor bearing mouse (n=5) received 3 se rial treatments of OXi4503 administered every 48 h. Figure 4 1 shows a typical 4T1 tumor treatment progression. Prior to OXi4503 administration, the tumor vessels as well as the normal vasculature were intact and functional with ample oxygenated blood supp ly ( Figure 4 1A, B). At 2 h post OXi4503 treatment, structural collapse of the tumor vasculature was observed ( Figure 4 1E) accompanied with significant reduction in oxygenation ( Figure 4 1F) with
66 initiation of minor vessel breakdown in some areas of the t umor. Progressive vessel disintegration advanced from the tumor core towards the periphery between 2 8 h post treatment resulting in functional loss in the tumor core (disintegrated vascular network highlighted in Figure 4 1G, H). Some tumor vessels did no t exhibit permanent structural alterations and instead temporarily lost functionality (example vessel highlighted in Figure 4 1I, J). As a result of the structural breakdown, the oxygenation in the tumor vasculature plummeted rapidly up to 6 h after treatm ent. From 8 48 h post treatment, a rapid structural and functional recovery ensued, progressing from the tumor periphery towards the tumor core. The vessels which had temporarily lost functionality without losing their structure were gradually reoxygenated ( Figure 4 1I, J O, P). The collapsed tumor vessels re emerged and reperfused ( Figure 4 1M, N O, P) whereas the vessels in which the structure was completely disintegrated did not recover ( Figure 4 1M, O). Rapid neovascularization and vascular remodeling w as observed between 24 48 h after OXi4503 administration in the form of new vessel sprouts and new network interconnections. Figure 4 2 gives a magnified view of the vessel remodeling and neovascularization in the tumor periphery during the recovery phase from 8 48 h after a treatment. Regional quantification of the vessel hemoglobin saturation was performed by dividing the 2 D image vasculature into tumor core, tumor periphery, or normal tissue regions. For example, the tumor nodule is indicated enclosed b y the dotted line in Figure 4 1A. Vessels outside the dotted line were considered as normal tissue vasculature. The tumor periphery vessels consisted of the vessels in the 250
67 thin border on the circumference of the tumor nodule. The remaining tumor vessels constituted the tumor core vasculature. Figure 4 3A shows the tumor before and after each of the serial treatments. An increasing fraction of vasculature deprived and apparently necrotic tumor area was observed with every serial treatment. In Figu re 4 3B is a cumulative plot for three serial treatments in five mice showing the HbSat changes over time. Two individual treatments from two different mice have been omitted from the data in Figure 4 3B owing to lack of considerable oxygenation changes. A s seen in Figure 4 3B, the core vessels experienced the greatest impact of the treatment followed by the periphery and the normal vessels. Interestingly, the adjacent normal vasculature also experienced oxygenation changes even though it remained structura lly intact throughout the treatment. From 2 6 h there was a steep decrease in all three region vessels with an onset of steady recovery from 8 h onward. The normal and tumor periphery vessels seemed to have an earlier and almost complete recovery back to t he original oxygenation values, as opposed to the tumor core vessel oxygenation. At 24 and 48 h post treatment normal and tumor periphery regions were highly oxygenated but the tumor core did not recover as much and remained relatively less oxygenated. Int erestingly, in one mouse the first of the three serial treatments was unable to elicit noticeable structural alterations with only subtle oxygenation changes, while in another mouse the second of the three serial treatments failed to produce any observable morphological or functional changes. Moreover, the later treatments in both these mice were able to produce profuse vascular damage with vivid changes in oxygenation.
68 No microvessel structural or functional changes were observed in the control mice admini stered carrier only treatment. In controls the tumor mass and vasculature continued to develop unrestrained. Over the entire experimental period, the tumor core region gradually becomes less oxygenated compared to the tumor periphery and normal tissue regi ons as the tumor expands. Figure 4 3C shows the hemoglobin saturation quantification results for carrier only treatments in 3 mice. Treatment of Caki 1 Human Renal C arcinomas An example OXi4503 treated Caki 1 tumor is shown in Figure 4 4. A pattern of str uctural and functional changes including vascular collapse, vessel disintegration, and oxygenation changes after OXi4503 treatment were observed. Notably, compared to 4T1 tumors the Caki 1 tumors manifested more vessel collapse and lesser vascular disinteg ration. Among the total number of treatments (n=18), only one single treatment in one mouse showed a comparable amount of tumor vessel disintegration to 4T1 tumors. For the rest of the treatments, as shown in Figure 4 4E, F I, J, the tumor vasculature unde rwent a temporary vessel collapse from 2 4 h with minimal structural breakdown. The Caki 1 tumor vasculature recovered earlier with the tumor periphery gradually rejuvenated by 6 h post treatment ( Figure 4 4I, J). During the recovery period from 6 48 h pos t treatment, the collapsed vessels reappeared. Similar to 4T1 tumors, some Caki 1 tumor vessels remained intact but experienced dramatic oxygenation changes. Even though there was minimal permanent vessel damage in Caki 1 tumors as compared to 4T1 tumors, there were explicit changes in vessel hemoglobin saturations in tumor and normal vasculature as seen in Fig ures 4 4 and 4 5. The brightfield images before and after each of the serial treatments is shown in Figure 4 5A and the cumulative quantified vessel oxygenation during each of the serial treatments in
69 five mice is shown in Figure 4 5B. From Figure 4 5B it can be seen that the tumor core in this tumor type did not have a decrease in microvessel oxygenation during the disruption phase as large in magnitu de compared to 4T1 tumors. The core vessel oxygenation was tightly coordinated with the tumor periphery and normal vessel oxygenation, and recovered as much and as fast as the latter two. From the brightfield images in Figure 4 5A it can be seen that compa red to 4T1 tumors ( Figure 4 3A) there was minimal vessel disintegration and no obvious avascular regions. The control mice which received the carrier only treatment did not show any dramatic structural and functional changes as shown in Figure 4 5C. With t he increasing tumor mass, there was a gradient in the tumor vessel oxygenation from normal surroundings towards tumor core, with core vessels being the least oxygenated. Even though there was no apparent necrosis evident in Caki 1 tumors, the tumor size se emed to be controlled by OXi4503 treatment. Tumor size in terms of apparent tumor area was measured from the 2 D brightfield images using ImageJ software (ImageJ 1.37a, Wayne Rasband, NIH, USA) and statistical size comparison between control and OXi4503 tr eated tumors was performed. On average the untreated tumors were significantly larger at 48 h (4.91 mm2, p<0.03, two tailed t test: paired two sample for means) than the starting tumor size for the first sham treatment (3.44 mm2), and twice the starting si ze after a fourth sham treatment (7.23 mm2). In contrast, the OXi4503 treated tumors did not show any statistically significant increase from their original size at the start of treatment (p>0.36, two tailed t test: paired two sample for means).
70 Rapid Oxyg enation Changes a fter OXi4503 T reatment A mouse with a 4T1 tumor was imaged at 30 sec intervals for 20 min prior to drug administration and 60 min after drug administration for real time recording of the immediate microvascular effects of the drug. In thi s particular example ( Figure 4 6 ) significant effects began at about 21 min after administration. There was a clear reduction in oxygenation of the tumor periphery, core and normal draining venule. Only the feeding artery oxygenation remained relatively un changed.
71 Figure 4 1 Typical OXi4503 treatment progression in a 4T1 tumor bearing mouse. The brightfield images (left hand figures in each panel) show structural alterations in the vasculature through the course of the single treatment. The pproxi mate tumor border is indicated by the dotted line in (A). Vessels outside the dotted line are co nsidered normal tissue vasculature. The tumor periphery consisted of a 250 to 500 tumor vasculature except for the periphery constituted the tumor core. Images were obtained at x2.5 magnification with image dimensions of 4.15x3.125 mm. The HbSat maps are shown in the right hand figures in each panel. The oxygenation levels in the HbSat maps are color coded as indicated by the colorbar. Arrows in frames (G) and (H) highlight the disintegrating vasc ulature, while arrows in (I) and (J) indicate an example vessel that persists through all time points with only oxygenation changes. The star in frames (M) and (O) indicate the avascular regions created by the OXi4503 treatment.
72 Figure 4 2. Recovery, r e oxygenation, vascular remodeling, and neovascularization in the 4T1 tumor during the recovery phase after OXi4503 treatment for the area indicated in Figure 1O by the dotted rectangle. The approximate tumor border is demarcated by the dotted line in the first bright field image. The arrows highlight an example vessel that maintains its structure throughout the treatment but undergoes dramatic oxygenation changes. The color scale for hemoglobin saturation is the same as in Figure 4 1.
73 Figu re 4 3 Quantitative tumor regional analysis for 4T1 tumor treatment. The tumor was divided into core, periphery, and normal tissue for purposes of analysis. (A) Brightfield images before and after every successive treatment of the tumor in Figure 4 1. (B ) Cumulative HbSat values (mean standard error) for 3 serial treatments in all 5 4T1 tumor mice. Arrows indicate drug administration. (C) Cumulative HbSat values (mean standard error) for 3 serial sham treatments (saline) in all 3 control mice. A. B C
74 Figu re 4 4. OXi4503 treatment 2 of a Caki 1 tumor bearing mouse. Images were obtained at x2.5 magnification with image dimensions of 4.15x3.125 mm. The HbSat maps are shown in the right hand figures in each panel. The oxygenation levels in the HbSat maps are color coded as indicated by the colorbar.
75 Figure 4 5 Quantitative tumor regional analysis for Caki 1 tumor treatment. The tumor was divided into core, periphery, and normal tissue for purposes of analysis. (A) Brightfield images before and after every successive treatment of the tumor in Figure 6 4. (B) Cumulative HbSat values (mean standard error) for 3 serial treatments in all 5 4T1 tumor mice. Arrows indicate drug administration. (C) Cumulative HbSat values (mean standard error) for 3 serial sham treatments (saline) in a control mouse. A. B C
76 Figure 4 6. Rapid oxygenation changes in 4T1 tumor microvasculature after treatment with VDA OXi4503. The mouse was imaged for 20 min before treatment and for 60 min after treatment. Spectral ima ges were obtained every 30 sec. (A) Brightfield image of the tumor microvasculature. (B) Hemoglobin saturation before and after OXi4503 treatment. A. B
77 Discussion OXi4503, a Combretastatin family member, is a tubulin binding drug [161, 162] and currently considered among the mo st potent VDAs [27, 163] In this study, we employed spectral imaging of window chamber tumors to investigate the realtime VDA induced modulation in microvessel struc ture and function of two different tumor types during multiple serial treatments with OXi4503. Many of the observations with our imaging system were consistent with those reported by others using different techniques. For example, both 4T1 and Caki 1 tumor s experienced a rapid drop in vascular oxygenation followed by recovery over 48 h post treatment. The timings of our observations for vascular disruption and recovery are similar to those reported by Salmon and Siemann using Hoechst 33342 vessel staining f or studying vascular shutdown and dynamic contrast enhanced magnetic resonance imaging for quantifying tumor perfusion inhibition  Both these methods suggested 80 90% tumor perfusion reduction at 4 h, with onset of recovery from 24 48 h (~63% recovery at 48 h). In our study the Caki 1 tumor recovery started earlier than 4T1 tumor recovery beginning at about 6 h post treatment while 4T1 recovery started around 8 h post treatment. In previous studies using tumor histology to assess VDA effects, there were indirect indications of vessel hemorrha ging due to the presence of red blood cells in the tissue  With our system, direct real time visualization of vessel hemorrhaging was possible and confirmed by hemoglobin saturation measurements that showed the presence of deoxygenated extravasated red blood cells. We observed rapid onset of microvessel functional changes in the form of decreasing hemoglobin saturation occurring mostly in tumor capillaries and draining venules starting about 21 min after administration of OXi4503, consistent with reported times for other VDAs 
78 Real time functional imaging of serial OXi4503 treatments highlighted unexpected differences in the responses of the 4T1 and Caki 1 tumors. The Caki 1 tumors were less responsive to vascul ar damage induced by OXi4503 than 4T1 tumors. The major Caki 1 tumor response was temporary vessel collapse with minimal vascular disruption followed by reperfusion and reoxygenation at later time points during recovery. Only a small fraction of the tumor core vessels did not recover. This response was very reproducible with multiple serial treatments. In 4T1 tumors there was extensive permanent microvessel damage and more vessel damage occurred with each subsequent treatment. Consistent with the greater va scular breakdown in 4T1 tumors, there was greater deoxygenation in 4T1 core vessels than in Caki 1 tumors. Despite the fact that in the Caki 1 tumors there was more temporary vessel collapse than permanent vessel disruption, treatment with OXi4503 seemed t o control tumor growth similar to 4T1 tumors. It should be noted that while there was little permanent microvessel disruption and a lack of apparent necrosis in the Caki 1 tumors after OXi4503 administration, the treatments were highly dynamic and perturba tive in terms of the acute effects on microvessel oxygenation. Another difference in the tumor treatment response was the lack of apparent necrotic areas in the Caki 1 tumors even after repeat doses, while similarly sized 4T1 tumors developed what appeared to be extensive necrotic areas in the tumor core. The observed induction of necrosis in the 4T1 tumors was unexpected given previous research that showed size dependence in tumor response to VDA treatment. Siemann and Rojiani, using several different rode nt and human tumor xenografts including Caki 1 tumors in mice, showed that larger tumors (>1 g) had significantly larger areas of
79 necrosis on the order of 90% after treatment with ZD6126 than smaller tumors (<0.3 g, ~25% necrosis) treated with the same dos e  The window chamber tumors fall into the smaller tumor size category, so the lack of apparent necrotic areas in the Caki 1 tumors in this study is consistent with the previous results with ZD6126, but the response of the 4T1 tumors was unexpected at this tumor size. It has been rep orted that different tumors can have diverse responses to the same VDA treatment owing to inherent variations in their tumor microenvironments or vasculature. For example, upon administration of a single dose of CA4 P to breast adenocarcinoma CaNT tumors a nd the round cell sarcoma SaS, Parkins et al observed greater vascular damage in the tumors having higher nitric oxide synthase activity  While the functional vascular volume, assessed using a fluorescent carbocyanine dye, was significantly reduced at 18 h after CA4 P treatment in both tumor types, the degree of reduction was very different in the two tumors. Similarly, Skliarenko et al have reported that the KHT C murine fibrosarcoma cell line and the CaSki human cervical carcinoma cell line responded dif ferently to the same one time treatment with ZD6126 depending on their initial interstitial fluid pressures (IFP)  Notably, the two tumor types manifested IFP changes distinct from each other, and the tumors with higher IFP s had less tumor cell death. A single tumor can also have different responses to different VDAs. Dalal and Burchill treated TC 32 and A673 Ewing's sarcoma family of tumors with CA4 P, OXi4503, and OXi8007, and demonstrated that different VDAs can have vary ing efficacy and treatment results in a particular tumor type  In light of the previous results, it is no t unreasonable that 4T1 and Caki 1 tumors could vary in their response to OXi4503, and that the differential response with tumor size reported with ZD6126
80 could be somewhat dissimilar with OXi4503. Further research is required to identify the conditions an d mechanisms responsible for the differences in 4T1 and Caki 1 small tumor responses to OXi4503. In summary, we used spectral imaging of tumor microvessel hemoglobin saturation with mouse window chamber tumors to measure the real time response of 4T1 and C aki 1 tumors to repeat treatments with OXi4503. Some of the observations made with our imaging system were consistent with others in the literature obtained with different techniques, such as the timing of the onset of vascular changes induced after admini stration of the VDA. We also observed differences in the responses of the 4T1 and Caki 1 tumors. Specifically, the 4T1 tumors were more responsive to treatment with OXi4503 than Caki 1 tumors in terms of permanent vascular damage and induction of tumor nec rosis, a finding that contrasted with previous work using a different VDA. A better understanding of the mechanisms of action of VDAs and how they can vary across tumors or at different stages of tumor development may help optimize the application of these agents and improve their efficacy.
81 CHAPTER 5 OXI4503 AND AVASTIN COMBINATION TREATMEN T OF CAKI 1 TUMORS Introduction All tumors need to develop their own vasculature to grow beyond a size of approximately 1 mm 3 [170 ] Vasculature development by tumors require numerous triggering mec hanisms such hypoxia and autocrine growth factor loops among others. H ypoxia or oxygen deprivation is known to accelerate angiogenesis in tumors which in turn up regulates the vascular endothelial growth factor (VEGF). Increase in VEGF secretion results in angiogenesis which is an inherent characteristic of solid tumors. Although, other factors such as angiopoietins, platelet derived growth factor B (PDGF B) etc  contribute to angiogenesis, VEGF is the rate limiting step in normal and pathological blood vessel growth  Hence, inhibiting VEGF function is an attractive strategy in retarding tumor growth. Based on evidence of its angiogenesis inhibition potential via VEGF blocking, recently a drug called Avastin or Bevacizumab has been approved for clinical us e. Avastin is the humanized form of the mouse anti VEGF monoclonal antibody A.4.6.1 that has been utilized widely in research for studying different tumor models. It functions by binding and neutralizing all human VEGF A isoforms [35, 45] Although a n giogenesis inhibiting agents including Avastin by itse lf more often than not are found lacking to be a curative as a single therapeutic agent [3, 15] The complex mechanism of angiogenesis has many triggers, and t umor growth and metastasi s being dependent on angiogenesis, the merit of AIs mostly lies in disease control as opposed to tumor cell kill [23, 55] Furthermore, e vidence suggests that cytotoxic therapies
82 including radiation therapy can initiate tumor angiogenesis  Hence the need for combination of AIs with conventional treatment s such as radiation is justified. In addition, de spite their efficacy in terms of vascular damage and tumor cell kill, VDAs alone cannot completely block the nutrient and oxygen supply to t he tumor since, the vasculature along the periphery of the tumor is left intact Hence, combination of VDAs with other therapies is a logical step in order to achieve complete tumor therapy. It is well known and has been well characterized that the VDAs such as Oxi4503 utiliz e the ir tubulin binding capability to impart a strong antivascular and antitum or effect In fact, the effect of Oxi4503 when tested with the models of murine breast adenocarcinoma CaNT the murine colon tumor MAC29, the human adenocarcinoma MDA MB 231 and the murine myocardial endothelioma MHEC5 T, OXi4503 exhibits a stronger antiv ascular and antitumor effect than CA4P  Notably, Oxi4503 has been shown to retard tumor growth in a dose dependent manner and improved survival in murine model of colorectal liver metastases  Furthermore, Oxi4503 works by binding the tubulin subunits thus preventing the formation of microtubles, which in turn leads to rounding up of endothelial cells. This rounding up of cells ultimately leads to the collapse of vascular wall and formation of thrombus which blocks the blood suppl y to the tumor [3, 29] However, OX i4503 treatment in itself does not lead to complete destruction of the tumor tissue. Hence, a combination of growth retardation of vasculature aroun d the tumor via AI s such as Avastin and dis ruptio n of existing vasculature via VDA s such as Oxi4503 forms an attractive complimentary and combinatorial strategy for amelioration of solid tumors and warrants further investigation
83 Furthermore, it is impera tive to understand the microvessel oxygenation to evaluate the tumor response to different regimes of Oxi4503 and Avastin drug treatment not only because they induce oxygenation fluctuations, but also due to the relevance of tumor oxygen status for other t herapeutic modalities Spectral imaging provides direct information regarding microvessel oxygenation in terms of HbSat. Previously t umor oxygenation has been measured for characterization of AI drug treatment response using various techniques including Ep pendorf polarographic oxygen electrodes [60 62] various hypoxia markers [63 65] or radiation response assays [66 68] Moreover, VDA treatment evaluation also involves tumor microv essel assessment fo r structure as well as functional changes such as vessel perfusion, permeability, blood flow etc [28, 29, 63 71] S pectral imaging used with the window chamber model allows visualization of the tumor in a live animal over a period of time. The tumor microvessel oxygenation information in terms of blood hemoglobin sa turation along with vessel network structure changes can be studied serially with great temporal and spatial resolution. Such information is an asset when studying the effect of VTAs on tumors. Hence, this method was utilized for studying the combinatoria l approach on the tumor pathology. Specifically, we studied the effect of individual as well as combinatorial treatment of human renal carcinomas (Caki 1 tumors) with two different VTAs (OXi4503, a tubular binding vascular disrupting agent; and Avastin, an anti VEGF angiogenesis inhibitor) using our mouse window chamber model and spectral imaging system. The eventual goal of this research is the in depth characterization and optimal combinatorial scheduling of VTAs with conventional therapies.
84 Materials a nd M ethods Tumor C ells Caki 1 human derived renal carcinoma xenografts were used for the experiments described in this chapter. Caki 1 tumor xenografts  were initiated in a single hind limb of donor mice by inoculating with 2 10 5 to 5 10 5 tumor cells. Tumors were excised when they reached about 200 mm 3 in size, and a single cell suspension was prepared in saline About 1 10 6 cells were used to initiate a window chamber tumor in the experimental mice. Animal M odel Details about the mouse model have been discussed in previous chapters. All i n vivo procedures were conducted in accordance with a protocol approved by University of Florida Institutional Animal Care and Use Committee. Athymic (nu/nu) female mice weighing at least 21 g (Charles River Laboratories, Raleigh, NC) were surgically impla nted with a titanium window chamber on the dorsal skin flap. During surgery mice were anesthetized by an IP injection of ketamine (100 mg/kg) and xylazine (10 mg/kg). A tumor was initiated in the window chamber during surgery by injecting the tumor cells ( 10 6 cells in 10 20 l saline ) subcutaneously in the dorsal skin flap prior to placing a 12 mm diameter no. 2 round glass coverslip (Erie Scientific, Portsmouth, NH) over the exposed skin. After surgery animals were housed in an environmental chamber maintai 12 h light/dark cycles.
85 Drug Preparation and Experimental D esign Vascular disrupting a gent OXi4503 OXi4503 a gift from OXiGENE, Inc. (Waltham, MA) was administered at a dose of 1 0 mg/kg IP (0.01 ml/g of mouse body weight). The light sensitive drug was protected from light during handling. OXi4503 was suspended in a solution of sterile saline and 50 f resh daily. Control groups received an equivalent volume of sterile saline and 5% sodium carbonate solution. Angiogenesis i nhibitor Avastin Avastin was received from Siemann lab. The stock solution was stored in 4 C and diluted to a dose concentration of 2 mg/Kg (0.01 ml/g of mouse body weight). Controls received equivalent amount of saline. Individual and c ombinatorial treatment d esign and imaging s chedule The Caki 1 tumors were allowed to grow for 8 12 days before starting the treatment. For individual t reatment with OXi4503, the treatment was repeated every 48 hours for up to 3 times per mouse (Given on e.g. MWF of the week) Spectral imaging was performed immediately prior to the treatment, and also at 4, 8, 24 and 48 h post treatment. After the 48 h im aging time point the mouse was re treated and the same imaging schedule was maintained In all, 12 mice bearing Caki 1 tum ors were treated with OXi4503 and 3 mice were carrier only controls. For individual treatment with Avastin, the treatment was given t wice each week (Given on e.g. M and F of the week) Mice that received combination treatment, Avastin was always injected 1 hour post OXi4503 administration. Imaging time points were
86 maintained same as the OXi4503 schedule for all individual, combination, and control treatments In all, 12 mice received OXi4503 only. Combination treatment was given to 5 mice. Three mice recei ved Avastin only, whereas 3 mice were controls receiving carrier only (saline or saline and 5% sodium carbonate solution) Tumor Size Measurement Tumors were measured with calipers everyday of imaging and tu mor volume was calculated assuming that the tumor volume as a half of an ellipsoid. The following equation was used to calculate tumor volume in mm 3 : Where, major, minor and vertic al represent the three mutually perpendicular axes of the ellipsoid. Imaging System, Image Acquisition and Image P rocessing The spectral imaging system, image acquisition, and image processing methods for hemoglobin saturation measurements were discussed in detail previously  Briefly, a Zeiss AxioImager microscope (Carl Zeiss, Inc., Thornwood, NY) was used as a basic platform for the imaging system. For transillumination images, a 100 W tungsten halogen lamp was used. The long working distance objectives used were 2.5 and 5 Fluars, 10 EC Plan NeoFluar and 20 LD Plan NeoFluar (Carl Zeiss, Inc.). For spectral information, bri ghtfield transmitted light images were obtained using a CCD camera thermoelectrically cooled to T2 FW). Band limited optical filtering was accomplished using a C mounted liquid crystal tunable filter (CRI, Inc., Woburn, MA) with a 400 to 720 nm transmission range and a
87 10 nm nominal bandwidth, placed in front of the CCD camera. LabVIEW8 (National Instruments Corp., Austin, TX) was used to automatically control the tuning of the LCTF and acquire images with the CC D camera. Microvessel hemoglobin saturation measurements and images were created from the spectral image data [90, 129] Statistical Analysis The least mean squares values and standard error were obtained using two way ANOVA via General Linear Model program, SYSTAT 12 (Version 12, Systat Software, San Jose, CA), with independent variables being the time points and treatment number. significant difference test. The data obtained from at least 3 different mice was pooled together to perform the statistics. Results Tumor Model I dentification Caki 1 t umor g rowth after individual t reatment with OXi4503 From our previous work with OXi4503, we observed that compared to 4T1 mouse mammary carcinomas, Caki 1 human derived tumors consistentl y appeared less susceptible to permanent vascular damage in response to the VDA (Figure 4 4 and 5 from Chapter 4 ). We found out that several studies so far have suggested a link between tumor size and VDA response [25, 166, 171] (from proposal) A recent study by Masunaga et al also stressed the importance of tumor size along with time of dos age for optimal response in individual ZD6126 treatment as well as when using it in combination with the cytotoxins tirapazamine (TPZ) and gamma rays  Siemann et al have specifically shown to tumor size dependence of Caki 1 tumors to another VDA named ZD6126  This fact, in addition to our observations of profuse vascular damage wi th
88 our small sized 4T1 tumors (~3mm diameter) suggested that larger Caki 1 tumors may be able to induce better response for OXi4503 treatments. Thus, Caki 1 tumors were grown in different sizes and treated with 1 3 serial treatments every 48 hours with th e VDA OXi4503 As a measure of treatment response, tumor volume measurement, HbSat changes, and vascular changes were monitored. Figure 5 1 shows the data from tumor volume comparison after OXi4503 treatments. It was observed that OXi4503 treated tumors up to 50 mm 3 volume showed greater tumor size control as compared to controls in our mouse model. Figure 5 1. Effect of OXi4503 treatment on different sized Caki 1 tumors. As shown in Figure 5 1A (i) and (ii), t umors with pre OXi4503 volumes up to 50 mm 3 demonstrated c ontrol of tumor growth compared to controls in the window chamber model
89 Effect of OXi4503 treatment on Caki 1 tumors: H emoglobin saturation (HbSat) a ssessment The tumors were serially treated with OXi4503 every 48 hours and HbSat maps of tumors and normal vasculature were obtained. Figure 5 2 demonstrates an example treatment with OXi4503. At the time of treatment the tumor was approximately 55 mm 3 in volume and 12 days old. At each imaging time point multiple images of the tumor and adja cent vasculature were obtained and complete tumor image was obtained by collage of several regions. As evident from the data, there was extensive and gradual structural damage and function loss in the t umor vasculature up to 8 h upon OXi4503 treatment. Th e vessels appeared to have constricted in diameter at 4 h (Figure 5 3 Brightfield image 4h post Tx) along with complete cessation of oxygenated blood supply (Figure 5 3 HbSat image 4h post Tx). The tumor periphery vessels start regaining oxygenated supply from 8 h onward, whereas the regrowth of tumor peripheral vessels appears to have started in the wake of oxygenation recovery. The tumor core lost bulk va scular network which did not recover and le d to massive avascular regions in the tumor body likely cau sing tumor cell death. The evidence of effective treatment was also seen in the form of tumor size reduction at 4 8 h time point from 55 to approximately 42 mm 3 for this particular tumor.
90 Figure 5 2 Example of one of the serial OXi4503 treatments in one mouse. The brightfield images are shown along with the corresponding hemoglobin saturation maps At t he beginning of treatment the tumor was approximately 55 mm 3 in volume and 12 days old. Structural deterioration of tumor vessel network is evident in the brightfield images up to 8 h post Tx, with recovery of peripheral vessels from 24 h. Oxygenation plummeted corresponding to the structural damage nearly to 8 h and recovered on the tumor rim from 8 h. At 48 h post Tx tumor had shrunk to nearly 42 mm 3 i n volume
91 Combination Treatment with OXi4503 and Avastin Tumor growth s tudy The data in figure 5 3 shows the comparison of tumor volumes after they were subjected to serial treatments with either OXi4503 or Avastin alone, or with a combination of the two. It was observed that even though OXi4503 treatment alone was able to effectively control tumor size (Figure 5 3 C), the combination treatment was even more effective in obliterating the tumors (Figure 5 3D) as compared to the individual treatments with eith er of the VTAs. A lthough A vastin treatment did slightly slow down the tumor growth as compared to controls in the one week of Avastin treatment (Figure 5 3A and B) its effect was very limited as compared to OXi4503 or combination treatments. Statistically significant pairs are indicated in Figure 5 3E.
92 Figure 5 3 Tumor volume comparison after individual and combination treatment with OXi4503 and Avastin. It was observed that combination of therapeutics have additive/supra additive effects on tumor si ze reduction of Caki1 tumors. A) The carrier treatment had no effect on tumor growth retardation. B) Avastin alone did not reduce the tumor size appreciably. C) The treatment of OX i 4503 alone retarded the tumor growth D) Trend suggests that the combinator i al treatment of Avastin and OXi4503 reduces the tumor size with each consecutive treatment E) Interesting significant ly different pairs (p<0.05) are as follows: treatments marked with as compared to 48h post T x 2 and Tx3 of Avastin
93 Quantitative HbSat a nalysis and c omparison Oxygenation change is a major characteristic of VDA treatment. AI treatments are also reported to have oxygenation ch anges. Hence we quantified and compared HbSat values for all the VTA treatments Four different treatment regime n s w ere undertaken to compare the oxygenation of the tumor microvasculature: 1) OX i4503 only treatment, 2) Avastin only treatment, 3) C ombination of Avastin and Oxi4503 and 4) Carrier only control treatment. Quantitative microvessel oxygenation was obtained i n terms of blood HbSat using methods previously described. Briefly, oxy and deoxy hemoglobin have distinct photo absorbance spectrum between 500 575 nm. Spectral brightfield information was obtained in the absorbance range of blood hemoglobin and HbSat va lues were derived using linear least square regression model  For assessment of oxygenation changes in various parts of the tumor, tumor vasculature was classified as belonging to the tumor core periphery, or the normal
94 vasculature outside the tumor (Figure 5 4) Figure 5 4. Example of classification of tumor core, periphery and normal vasculature. The periphery was chosen between 250 500 m on the tumor rim. For analysis, the HbSat values obtained were first normalized with respect to the values ob tained for the time point just before (0 h) treating the mice with different conditions. In all, 12 mice were treated with Oxi4503 only treatment, 5 mice were treated with Avastin only treatment, whereas 5 mice were treated with combination treatment of Av astin and Oxi4503 and 3 mice were utilized for control by delivery of carrier only. Hyperspectral images were obtained and HbSat values were quantitatively determined at different time points of 0h, 4h, 8h, 24h, and 48h after each treatment. As many as 3 t reatment regime n s were provided for each individual treatment type. Data Core vessels Periphery Normal vasculature 500
95 honestly significant differences between oxygenation levels at different time points. The average va lues for different time points with standard error are plotted for all the treatment types ( Figure 5 5 ). Upon treatment of Oxi4503, it was observed that after 4 h, the oxygenation in the periphery, core and normal vessels dipped by 20 5 0%, which was found to be significantly different from the initial oxy genation levels at 0 h (Figure 5 5 Ai, Aii, Aiii). It is evident from the plots that even though the peripheral and normal vessels underwent significant differences in oxygenation after each treatment, the ir recovery was faster than the core vessels that were affected more and regained oxygenation slower after each OXi4503 treatment. Notably, in the periphery vasculature for the first and second treatment, 4 h oxygenation level was significant ly different w ith all the other time points for the respective treatments. Furthermore, it was observed that there were significant differences between 4 h and 8 h for both first and second treatment in the normal vasculature. Additionally, there was significant differe nce between 4 h and 48 h oxygenation levels for the second treatment in the normal vasculature. However, in case of core vessels, 4 h oxygenation levels were statistically different from 0 h and 48 h oxygenation only. (Figure 5 5 Ai, Aii, Aiii). For Avasti n only treatment, in the periphery vasculature the oxygenation level was found to be significantly improved at 72 h post Tx 1 as compared to 8 h and 24 h There were no significant differences found for the core and normal vasculature f or this treatment typ e (Figure 5 5 Bi, Bii, Biii) although an overall incremental trend in the oxygenation was observed after Avastin treatments
96 Interestingly, for the combination treatment, significant differences were observed in the second treatment group between oxygena tion levels at 4 h and 0 h; and 4 h and 48 h for the periphery vasculature. Contrarily, although, for the core vasculature, significant differences were not observed in the second treatment there were significant differences observed in the first treatment between oxygenation levels of 0 h and 4 h; 0 h and 8h; 4 h and 48 h. Additionally, for the normal vasculature, although, there was significant difference between oxygenation levels at 8 h and 48 h, the change in oxygenatio n values was > 20% (Figure 5 5 Ci Cii, Ciii). As expected, the control of carrier only vasculature did not have any significant differences between oxygenation levels at different time points for periphery, core and normal vasculature (Figure 5 5 D).
97 Figure 5 5 The HbSat or oxygenation of the periphery, core and normal microvessels modulates with individual trea tments with (A) Oxi4503 or (B) A vastin, or their (C) combination at different t ime points and (D) control. Two way ANOVA was performed on HbSat valu es normalized to 0 h time points of each treatment obtained from more than 3 mice per group and the average values of HbSat and standard error are plotted against corresponding time points. Significant differences between individual time points are indicat ed. The black arrows represent the time point where treatment was given to mice. HbSat (Normalized) Time (Hours) Time (Hours) Time (Hours) Periphery Core Normal (A) Oxi4503 Treatment Only * (Ai) ( Aii) (Aiii) HbSat (Normalized) (D) Control carrier only HbSat (Normalized) Normal Time (Hours) Periphery (B) Avastin Treatment Only Core Time (Hours) Time (Hours) (Bi) (Bii) (Biii) HbSat ( Normalized) Normal (C) Combinatorial Approach Periphery Core * (Ci) (Cii) (Ciii) Time (Hours) Time (Hours) Time (Hours) Time (Hours)
98 Discussion Neoplasms need neovasculature to develop beyond a few millimeters in size  Tumors employ several differ ent mechanisms for inducing angiogenesis thereby causing endothelial cell proliferation and sprouting of new microvessels from the existing vasculature [10, 18, 19] Tumor vasculature makes a lucrative target for drug development owing to the cru cial role of endothelial cells in cancer cell survival, growth, and metastasis, and because they are more genetically stable, and potentially less likely to develop resistance to therapeutic agents as compared to tumor cells themselves  Moreover, targeting endothelial cells offers new avenues for therapeutic targeting than traditional chemotherapeutics, thereby implying potential dual benefit from combination approach. A number of strategi es a re in development for targeting tumor vasculature. Some approaches focus on inhibiting neo vasculature sprouting by blocking the angiogenic pathways (Angiogenesis inhibitors, AIs), while others seek to destroy the existing tumor vasculature (Vascular d isrupting agents, VDAs)  Combination of these two mechanisms to target tumor vasculature is a new approach being currently investigated. It has been observed that a n giogenesis inhibiting agents more ofte n than not are found lacking to be a curative as a single therapeutic agent [3, 15] The complex mechanism of angiogenesis has many triggers, and t umor growth and metastasis being dependent on angiogenesis, the merit of AIs mostly lies in disease control as opposed to tumor cell kill [23, 55] In addition, de spite their efficacy in terms of vascular damage and tumor cell kill, VDAs alone canno t completely block the nutrient and oxygen supply to t he tumor since, the vasculature along the periphery of the tumor is left intact
99 Combination of AIs with VDAs is under investigation and shows promise for comprehensive tumor therapy [24 26] Due to their complimentary mechanisms of action a nd non overlapping treatment schedules, AIs and VDAs may be used together for enhanced tumor treatment as compared to individual treatment with these agents. Based on this synergistic mechanism hypothesis, using spectral imaging techniques we endeavored t o evaluate the efficacy of c ombinat ion treatment with t wo of the popular VTAs, OXi4503 and Avastin in the Caki 1 renal cell carcinomas Human derived Caki 1 cells were chosen for the combination study since Avastin is a recombinant humanized monoclonal an tibody against vascular endothelial growth factor (VEGF)  Also, r enal cell carcinomas have been reported to have particularly high VEGF expressions potentially making them particularly susceptible to Avastin. OXi4503, the dipho sphate prodrug of CA1P  is a tubuli n binding agent currently regarded as one of the most potent VDAs [27, 163] It is a fast acting drug, shown to induce time dependent decrease in tumor perfusion, inc rease in tumor vessel permeability, and profuse tumor vascular damage in several tumor models [3, 27, 28] The anti angiogenic Avastin (Bevacizumab) is a recombi nant humanized monoclonal antibody against vascular endothelial growth factor (VEGF)  Owing to its significance for the process of angiogenesis, VEGF inhibition leads to arrest of tumor growth and metastasis  The non overlapping effects of these two VTAs, have been shown to improve tumor treatment measured in terms of incre ased growth delay, when used as a combination as opposed to individual treatment  Before using them for VTA comb ination treatments, preliminary characterization of Caki 1 tumors was performed. From our previous work with OXi4503  we
100 observed that compared to 4 T1 mouse mammary carcinomas (approximately 3 4mm in diameter) the Caki 1 human derived renal tumors (approximately 2 3mm in diameter) consistently appe ared less susceptible to permanent vascular damage in response to the VDA. We found out that several studies so far have suggested a link between tumor size and VDA response [25, 166, 171] This fact, in addition to our observations of profuse vascular damage with our small sized 4T1 tumors, indicated that larger Caki 1 tumors may induce bet ter response for OXi4503 treatments. Thus, Caki 1 tumors were grown at larger sizes (approximately 3.5 7 .5 mm in diameter) in the window chamber mouse model (approximately 1cm in diameter) compared to previous experiments (approximately 2 3mm in diameter) treated serially with OXi4503 up to 3 times, and their response evaluated. Our tumor growth data indicated that tumors less than or around 5mm in diameter (approximately 50 mm 3 in volume) demonstrated effective tumor growth retardation in the window chambe r model (Figure 5 1) Also, as seen in Figure (5 2), profuse vascular disintegration and oxygenation changes similar to previous observations in 4T1 tumors (Figure 4 1 and 3) further signified the suitability of Caki 1 tumors for studies with the combinati on treatments. Once the tumor model to be used for combination studies with OXi4503 and Avastin w as identified Caki 1 tumors were subjected to f our different treatment regime n s to compare the tumor growth retardation, and changes in structure and oxygenat ion of the tumor microvasculature: 1) OXi4503 only treatment, 2) Avastin only treatment, 3) Combination of Avastin and Oxi4503 and 4) Carrier only control treatment. Tumor growth monitoring post VTA treatments clearly revealed differences between the effi cacies of the individual and combination treatments (Figure 5 3).
101 Compared to the controls, Avastin only treatment marginally retarded the tumor growth rat e with one week of treatment (Figure 5 3A and B). This was somewhat expected since anti angiogenics such as Avastin are considered long term treatments, believed to cause tumor growth retardation over much longer periods. Although Avastin treatment did retard the tumor growth, its effect was very limited as compared to OXi4503 or combination treatments. Interestingly, even though OXi4503 treatment alone was able to effectively control tumor size (Figure 5 3C) the combination treatment was even more effective in obliterating the tumors (Figure 5 3D) as compared to the individual treatments with either of the VTAs thus confirming the findings of Siemann et al  Spectral imaging enabled us to obtain microvessel oxygen ation values in terms of hemoglobin saturation As depicted in Figure 5 5, this vessel oxygenation data evaluation and comparison between the various treatment regimens offered further insights into the mechanism of these drugs. The OXi4503 treatment induc ed consistent vascular structure and corresponding function loss at approximately 4 h after each of the serial treatments. With each serial treatment, the tumor core vessels suffered the most with loss of oxygenation up to nearly 50% of the pre treatment v alue, followed by tumor peripheral vessels ( approximately 30% oxygenation loss) and adjacent normal vasculature (approximately 20% oxygenation loss) (Figure 5 5Ai, ii, iii). The periphery and normal vessels recovered faster at the 8, 24 and 48 h post trea tment time points, whereas the core recovered much slower owing to the OXi4503 induced vascular network disruption. These results were similar to the results obtained with 4T1 tumors (Figure 4 3).
102 Changes in tumor vessel oxygenation post Avastin were subtl e compared to OXi4503 treatment. Though not statistically significant, overall incremental trend was observed in the normal, peripheral, and tumor core vasculature after 72 h post first Avastin treatment. Interestingly, for combination treatment the initia l treatments appear to induce oxygenation changes similar to individual OXi4503 treatment, with later serial treatments showing lesser fluctuations (Figure 5 5 Ci, Cii, Ciii). This makes us speculate the impact of Avastin on the treatment response. The mec hanisms responsible for such influence are still unknown and require further investigation. In summary, we identified an appropriate tumor model for VTA combination treatment evaluation and used spectral imaging of tumor microvessel hemoglobin saturation w ith mouse window chamber tumors to measure the real time response of Caki 1 tumors Some of the observations made with our imaging system support the literature as well as our previous work We were able to identify differences in the responses of Caki 1 t umors to individual and combination treatments with OXi4503 and Avastin. A better understanding of the mechanisms of action of V T As and how they can be used in synergy with each other may help optimize the application of these agents and improve their effi cacy.
103 CHAPTER 6 CONCLUSION AND FUTUR E DIRECTION In conclusion, we were able to utilize spectral imaging methods in live mouse window chambers to study the efficacy of novel tumor therapeutic drugs. We identified an appropriate tumor model for the evaluati ons of two leading vascular targeting agents namely OXi4503 and Avastin. Our findings clearly suggest that the combination of these two vascular targeting agents induce better therapeutic outcomes in terms of tumor growth retardation as compared to indivi dual treatments with either agent. Additionally, OXi4503 modulates the tumor microvascular response in the presence of Avastin Interestingly, in the later treatments, there is minimal permanent vessel damage along with frequent vascular collapse of the tu mor core vessels. We observe that t he se vessels recover from their collapsed state with rejuvenated blood supply and ample oxygenation. In the combination therapy, d espite minimal vessel disint egration, the tumors experience size reduction indicating necro sis or apoptosis of tumor cell s The mechanism s involved causing t he s e effect s are not yet known and warrant further investigation s The modulation of vascular response upon combination treatment might be explained by the varying VEGF production and VEGF blocking process that occurs during the combination treatment. Upon use of Avastin as a single agent reduction in neo angiogenesis of tumor vessels was observed however its effect appeared more dramatic when used in combination with the vascular disrupti ng agent Oxi4503 In the presence of Avastin, VEGF may have been blocked effectively to mellow down the recovery phase of OXi4503. Additional VEGF expression, along with the pathological tumor cell expression of VEGF, with OXi4503 treatment may have incre ased the
104 potency of Avastin. In order to dispel these numerous uncertainties, f urther studies to assess the functional VEGF activity prior to and after these VTA treatments are required Again, despite the apparent reduction in the vascular damage, the dat a suggests that tumor cell death may have been increased leading to the tumor size reduction. M echanisms responsible for this phenomenon need to be investigated. A better understanding of the mechanisms of action of VTAs and how they can be used in synerg y with each other may help optimize the application of these agents and improve their efficacy.
105 LIST OF REFERENCES 1. Siemann DW, Chaplin DJ: An update on the clinical development of drugs to disable tumor vasculature Expert Opin Drug Discov 2007, 2 (10):1 11. 2. Hirst DG, Flitney FW: The physiologic importance and therapeutic potential of nitric oxide in the tumour associated vasculature In: Tumour Angiogenesis. Edited by Bicknell R, Lewis CE, Ferrara N. Oxford: Oxf ord University Pre ss; 1997: 153 167. 3. Salmon HW, Siemann DW: Effect of the second generation vascular disrupting agent OXi4503 on tumor vascularity Am Assoc Cancer Res 2006, 12 (13):4090 4094. 4. Bikfalvi A: Angiogenesis: health and disease Ann Oncol 2006, 17 (Suppl 10) : 65 70. 5. Siemann DW: Vascular targeting agents Horizons in cancer therapeutics 2002, 3 :4 14. 6. Horsman MR, Siemann DW: Pathophysiologic effects of vascular targeting agents and the implications for combination with conventional therapies Cancer Res 2006, 66 (24):11520 11539. 7. Folkman J, Shing Y: Angiogenesis J Biol Chem 1992, 267 (16):10931 10934. 8. Arap W, Pasqualini R, Ruoslahti E: Cancer treatment by targeted drug delivery to tumor vasculature in a mouse model Science 1998, 279 :377 380. 9. Ruoslahti E: Specialization of tumor vasculature Nat Rev Cancer 2002, 2 :83 90. 10. Ellis LM, Liu W, Ahmad SA, F F, Jung YD, Shaheen RM, Reinmuth N: Overview of angiogenesis: Biologic implications for antiangiogenic therapy Semin Oncol 2001, 28 :94 104. 11. Kerbel RS: Tumor angiogenesis: past, present and the near future Carcenogenesis 2000, 21 :505 515. 12. Chaplin DJ, Dougherty GJ: Tumor vasculature as a target for cancer therapy Br J Cancer 1999, 80 (Suppl 1):57 64. 13. Denekamp J: The tumor microcir culation as a target in cancer therapy: a clearer perspective Eur J Clin Invest 1999, 29 :733 736. 14. Baguley BC: Antivascular therapy of cancer: DMXAA Lancet Oncol 2003, 4 :141 148.
106 15. Siemann DW, Chaplin DJ, Horsman MR: Vascular targeting therapies f or treatment of malignant disease Cancer 2004, 100 :2491 2499. 16. Thorpe PE: Vascular targeting agents as cancer therapeutics Clin Cancer Res 2004, 10 :415 427. 17. Siemann DW, Bibby MC, Dark GG, Dicker AP, Eskens FALM, Horsman MR, Marme D, LoRusso PM: Differentiation and definition of vascular targeted therapies Am Assoc Cancer Res 2005, 11 :416 420. 18. Marme D: The impact of anti angiogenic agents on cancer therapy J Cancer Res Clin Oncol 2003, 129 :607 620. 19. Kerbel R, Folkman J: Clinical transla tion of angiogenesis inhibitors Nat Rev Cancer 2002, 2 :727 739. 20. Sridhar SS, Shepherd FA: Targeting angiogenesis: a review of angiogenesis inhibitors in the treatment of lung cancer Lung Cancer 2003, 42 (Suppl 1):S81 91. 21. Eskens FA: Angiogenesis i nhibitors in clinical development; where are we now and where are we going? Br J Cancer 2004, 90 :1 7. 22. Wachsberger P, Burd R, Dicker AP: Tumor Response to Ionizing Radiation Combined with Antiangiogenesis or Vascular Targeting Agents: Exploring Mechani sms of Interaction Clinical Cancer Research 2003, 9 :1957 1971. 23. Siemann DW, Shi W: Targeting the Tumor Blood Vessel Network to Enhance the Efficacy of Radiation Therapy Seminars in Radiation Oncology 2003, 13 (1):53 61. 24. Siemann DW, Shi W: Dual ta rgeting of tumor vasculature: Combining Avastin and vascular disrupting agents (CA4P or OXi4503) Anti Cancer Research 2008, 28 :2027 2032. 25. Landuyt W, Ahmad B, Nuyts S, Theys J, Op dBM, Rijnders A, Anne J, van Oosterom A, van den Bogaert W, Lambin P: I n vivo antitumor effect of vascular targeting combined with either ionizing radiation or anti angiogenesis treatment Int J Radiat Oncol Biol Phys 2001, 49 :443 450. 26. Siemann DW, Shi W: Efficacy of combined antiangiogenic and vascular disrupting agents in treatment of solid tumors Int J Radiat Oncol Biol Phys 2004, 60 (4):1233 1240.
107 27. Sheng Y, Hua J, Pinney KG, Garner CM, Kane RR, Prezioso JA, Chaplin DJ, Edvardsen K: Combretastatin family member OXi4503 induces tumor vascular collapse through the ind uction of endothelial apoptosis Int J Cancer 2004, 111 :604 610. 28. Chan LS, Malcontenti Wilson C, Muralidharan V, Christophi C: Alterations in vascular architecture and permeability following OXi4503 treatment Anti Cancer Drugs 2008, 19 :17 22. 29. Sal mon HW, Mladinich C, Siemann DW: Evaluations of vascular disrupting agents CA4P and OXi4503 in renal cell carcinoma (Caki 1) using a silicon based microvascular casting technique Eur J Cancer 2006, 42 :3073 3078. 30. Ehrmann RL, Knoth M: Choriocarcinoma Transfilter Stimulation of Vasoproliferation in Hamster Cheek Pouch Studied by Light and Electron Microscopy Journal of the National Cancer Institute 1968, 41 (6):1329 1333 31. Folkman J, Bach M, Rowe JW, Davidoff F, Lambert P, Hirsch C, Goldberg A, Hia tt HH, Glass J, Henshaw E: Tumor Angiogenesis Therapeutic Implications N Engl J Med 1971, 285 (21):1182 1188 32. Folkman J, Klagsbrun M: Angiogenic Factors Science 1987, 235 (4787):442 447. 33. Ferrara N: VEGF and the quest for tumour angiogenesis fac tors Nat Rev Cancer 2002, 2 (10):795 803. 34. Jain RK: Molecular regulation of vessel maturation Nature Medicine 2003, 9 (6):685 693. 35. Ferrara N, Gerber HP, LeCouter J: The biology of VEGF and its receptors Nat Med 2003, 9 :669 676. 36. Volm M, Kooma gi R, Mattern J: Prognostic value of vascular endothelial growth factor and its receptor Flt 1 in squamous cell lung cancer International Journal of Cancer 1997, 74 (1):64 68. 37. Yoshiji H, Gomez DE, Shibuya M, Thorgeirsson UP: Expression of vascular end othelial growth factor, its receptor, and other angiogenic factors in human breast cancer Cancer Research 1996, 56 (9):2013 2016. 38. Ellis LM, Takahashi Y, Fenoglio CJ, Cleary KR, Bucana CD, Evans DB: Vessel counts and vascular endothelial growth factor expression in pancreatic adenocarcinoma European Journal of Cancer 1998, 34 (3):337 340.
108 39. Tomisawa M, Tokunaga T, Oshika Y, Tsuchida T, Fukushima Y, Sato H, Kijima H, Yamazaki H, Ueyama Y, Tamaoki N et al : Expression pattern of vascular endothelial gro wth factor isoform is closely correlated with tumour stage and vascularisation in renal cell carcinoma European Journal of Cancer 1999, 35 (1):133 137. 40. Sowter HM, Corps AN, Evans AL, Clark DE, Charnock Jones DS, Smith SK: Expression and localization o f the vascular endothelial growth factor family in ovarian epithelial tumors Laboratory Investigation 1997, 77 (6):607 614. 41. Kim KJ, Li B, Winer J, Armanini M, Gillett N, Phillips HS, Ferrara N: Inhibition of Vascular Endothelial Growth Factor Induced Angiogenesis Suppresses Tumor Growth Invivo Nature 1993, 362 (6423):841 844. 42. Gerber HP, Kowalski J, Sherman D, Eberhard DA, Ferrara N: Complete inhibition of rhabdomyosarcoma xenograft growth and neovascularization requires blockade of both tumor and host vascular endothelial growth factor Cancer Research 2000, 60 (22):6253 6258. 43. Kuo CJ, Farnebo F, Yu EY, Christofferson R, Swearingen RA, Carter R, von Recum HA, Yuan J, Kamihara J, Flynn E et al : Comparative evaluation of the antitumor activity of antiangiogenic proteins delivered by gene transfer Proc Natl Acad Sci U S A 2001, 98 (8):4605 4610. 44. Holash J, Davis S, Papadopoulos N, Croll SD, Ho L, Russell M, Boland P, Leidich R, Hylton D, Burova E et al : VEGF Trap: A VEGF blocker with potent anti tumor effects Proc Natl Acad Sci U S A 2002, 99 (17):11393 11398. 45. Presta LG, Chen H, Oconnor SJ, Chisholm V, Meng YG, Krummen L, Winkler M, Ferrara N: Humanization of an anti vascular endothelial growth factor monoclonal antibody for the therapy of so lid tumors and other disorders Cancer Research 1997, 57 (20):4593 4599. 46. Lin YS, Nguyen C, Mendoza JL, Escandon E, Fei D, Meng YG, Modi NB: Preclinical pharmacokinetics, interspecies scaling, and tissue distribution of a humanized monoclonal antibody a gainst vascular endothelial growth factor Journal of Pharmacology and Experimental Therapeutics 1999, 288 (1):371 378. 47. Shima DT, Gougos A, Miller JW, Tolentino M, Robinson G, Adamis AP, Damore PA: Cloning and mRNA expression of vascular endothelial gr owth factor in ischemic retinas of Macaca fascicularis Investigative Ophthalmology & Visual Science 1996, 37 (7):1334 1340.
109 48. Ryan AM, Eppler DB, Hagler KE, Bruner RH, Thomford PJ, Hall RL, Shopp GM, O'Neill CA: Preclinical safety evaluation of rhuMAbVE GF, an antiangiogenic humanized monoclonal antibody Toxicologic Pathology 1999, 27 (1):78 86. 49. Margolin K, Gordon MS, Holmgren E, Gaudreault J, Novotny W, Fyfe G, Adelman D, Stalter S, Breed J: Phase Ib trial of intravenous recombinant humanized monocl onal antibody to vascular endothelial growth factor in combination with chemotherapy in patients with advanced cancer: Pharmacologic and long term safety data Journal of Clinical Oncology 2001, 19 (3):851 856. 50. Gordon MS, Margolin K, Talpaz M, Sledge G W, Holmgren E, Benjamin R, Stalter S, Shak S, Adelman DC: Phase I safety and pharmacokinetic study of recombinant human anti vascular endothelial growth factor in patients with advanced cancer Journal of Clinical Oncology 2001, 19 (3):843 850. 51. Cobleig h MA, Langmuir VK, Sledge GW, Miller KD, Haney L, Novotny WF, Reimann JD, Vassel A: A phase I/II dose escalation trial of bevacizumab in previously treated metastatic breast cancer Seminars in Oncology 2003, 30 (5):117 124. 52. Yang JC, Haworth L, Sherry RM, Hwu P, Schwartzentruber DJ, Topalian SL, Steinberg SM, Chen HX, Rosenberg SA: A randomized trial of bevacizumab, an anti vascular endothelial growth factor antibody, for metastatic renal cancer N Engl J Med 2003, 349 (5):427 434. 53. Kabbinavar F, Hur witz HI, Fehrenbacher L, Meropol NJ, Novotny WF, Lieberman G, Griffing S, Bergsland E: Phase II, randomized trial comparing bevacizumab plus fluorouracil (FU) leucovorin (LV) with FU/LV alone in patients with metastatic colorectal cancer Journal of Clinic al Oncology 2003, 21 (1):60 65. 54. Li L, Rojiani AM, Siemann DW: Preclinical evaluations of therapies combining the vascular targeting agent Combretastatin A 4 Disodium Phosphate and conventional anticancer therapies in the treatment of kaposi's sarcoma Acta Oncologica 2002, 41 (1):91 97. 55. Liang WC, Wu X, Peale FV, Lee CV, Meng YG, Gutierrez J, Fu L, Malik A, Gerber HP, Ferrara N et al : Cross species Vascular Endothelial Growth Factor (VEGF) blocking Antibodies Completely Inhibit the Growth of Human Tu mor Xenografts and Measure the Contribution of Stromal VEGF THE JOURNAL OF BIOLOGICAL CHEMISTRY 2006, 281 (2):951 961. 56. Koukourakis M, Giatromanolaki A, Sivridis E, Simopoulos K, Pissakas G, Gatter K, Harris AL: Squamous cell head and neck cancer: evid ence of angiogenic regeneration during radiotherapy Anticancer Res 2001, 21 :4301 4309.
110 57. Gerber HP, Ferrara N: Pharmacology and Pharmacodynamics of Bevacizumab as Monotherapy or in Combination with Cytotoxic Therapy in Preclinical Studies Cancer Res 2 005, 65 (3). 58. Siemann DW, Mercer E, Lepler S, Rojiani AM: Vascular targeting agents ehnance chemotherapeutic agent activities in solid tumor therapy Int J Cancer 2002, 99 :1 6. 59. Li L, Rojiani AM, Siemann DW: Preclinical Evaluations of Therapies Comb ining the Vascular Targeting Agent Combretastatin A 4 Disodium Phosphate and Conventional Anticancer Therapies in the Treatment of Acta Oncologica 2002, 41 (1):91 97. 60. Lee CG, Heijn M, di Tomaso E, Griffon Etienne G, Ancukiewicz M, Koi ke C, Park K, Ferrara N, Jain RK, Suit HD et al : Anti Vascular Endothelial Growth Factor Treatment Augments Tumor Radiation Response under Normoxic or Hypoxic Conditions CANCER RESEARCH 2000, 60 :5565 5570. 61. Gong H, Pottgen C, Stuben G, Havers W, Stusc hke M, Schweigerer L: ARGININE DEIMINASE AND OTHER ANTIANGIOGENIC AGENTS INHIBIT UNFAVORABLE NEUROBLASTOMA GROWTH: POTENTIATION BY IRRADIATION Int J Cancer 2003, 106 :723 728. 62. Teicher BA, Holden SA, Ara G, Dupuis NP, Liu F, Yuan J, Ikebe M, Kakeji Y: Influence of an anti angiogenic treatment on 9L gliosarcoma: Oxygenation and response to cytotoxic therapy Int J Cancer 1995, 61 (5):732 737. 63. Fenton B, Paoni S, Grimwood B, Ding I: DISPARATE EFFECTS OF ENDOSTATIN ON TUMOR VASCULAR PERFUSION AND HYPOXI A IN TWO MURINE MAMMARY CARCINOMAS Int J Radiation Oncology Biol Phys 2003, 57 (4):1038 1046. 64. Fenton B, Paoni S, Ding I: Pathophysiological Effects of Vascular Endothelial Growth Factor Receptor 2 Blocking Antibody plus Fractionated Radiotherapy on Mu rine Mammary Tumors CANCER RESEARCH 2004, 64 :5712 5719. 65. Bernsen HJJA, Rijken PFJW, Peters JPW, Bakker JH, Boerman RH, Wesseling P, van der Kogel AJ: Suramin treatment of human glioma xenografts; effects on tumor vasculature and oxygenation status Jo urnal of Neuro Oncology 1999, 44 :129 136. 66. Leith JT, Papa G, Quaranto L, Michelson S: Modification of the volumetric growth responses and steady state hypoxic fractions of xenografted DLD 2
111 human colon c arcinomas by administration of basic fibroblast g rowth factor or suramin Br J Cancer 1992, 66 :345 348. 67. Murata R, Nishimura Y, Hiraoka M: AN ANTIANGIOGENIC AGENT (TNP 470) INHIBITED REOXYGENATION DURING FRACTIONATED RADIOTHERAPY OF MURINE MAMMARY CARCINOMA Int J Radiation Oncology Biol Phys 1997, 3 7 (5):1107 1113. 68. Rofstad E, Henriksen K, Galappathi K, Mathiesen B: Antiangiogenic Treatment with Thrombospondin 1 Enhances Primary Tumor Radiation Response and Prevents Growth of Dormant Pulmonary Micrometastases after Curative Radiation Therapy in H uman Melanoma Xenografts CANCER RESEARCH 2003, 63 :4055 4061. 69. El Emir E, Boxer GM, Petrie IA, Boden RW, Dearling JLJ, Begent RHJ, Pedley RB: Tumour parameters affected by combretastatin A 4 phosphate therapy in a human colorectal xenograft model in nu de mice European Journal of Cancer 41 :799 806. 70. Galbraith SM, Maxwell RJ, Lodge MA, Tozer GM, Wilson J, Taylor NJ, Stirling JJ, Sena L, Padhani AR, Rustin GJS: Combretastatin A4 Phosphate Has Tumor Antivascular Activity in Rat and Man as Demonstrated by Dynamic Magnetic Resonance Imaging J Clin Oncol 2003, 21 :2831 2842. 71. Seshadri M, Spernyak JA, Maiery PG, Richard T. Cheneyz, Mazurchuk R, Bellnier DA: Visualizing the Acute Effects of Vascular Targeted Therapy In Vivo Using Intravital Microscopy an d Magnetic Resonance Imaging: Correlation with Endothelial Apoptosis, Cytokine Induction, and Treatment Outcome Neoplasia 2007, 9 (2):128 135. 72. Kotz B, West C, Saleem A, Jones T, Price P: Blood flow and Vd (water): both biomarkers required for interpr eting the effects of vascular targeting agents on tumor and normal tissue Mol Cancer Ther 2009, 8 (2):303 309. 73. Tozer GM, Akerman S, Cross NA, Barber PR, Bjo¨rndahl MA, Greco O, Harris S, Hill SA, Honess DJ, Ireson CR et al : Blood Vessel Maturation and Response to Vascular Disrupting Therapy in Single Vascular Endothelial Growth Factor A Isoform Producing Tumors Cancer Res 2008, 68 (7):2301 2311. 74. deDeugd C, Wankhede M, Sorg BS: Multimodal optical imaging of microvessel network convective oxygen tra nsport dynamics Applied Optics 2009, 48 (10):D187 D197. 75. Kimura H, Braun RD, Ong ET, Hsu R, Secomb TW, PapaHadjopoulos D, Hong K, Dewhirst M: Fluctuations in red cell flux in tumor microvessels can lead
112 to transient hypoxia and reoxygenation in tumor p arenchyma Cancer Res 1996, 56 :5522 5528. 76. Kerger H, Torres IP, Rivas M, Winslow RM, Intaglietta M: Systemic and Subcutaneous Microvascular Oxygen Tension in Conscious Syrian Golden Hamsters American Journal of Physiology Heart and Circulatory Physiol ogy 1995, 268 (2):H802 H810. 77. Menger MD, Laschke MW, Vollmar B: Viewing the microcirculation through the window: Some twenty years experience with the hamster dorsal skinfold chamber European Surgical Research 2002, 34 (1 2):83 91. 78. Laschke MW, Meng er MD: In vitro and in vivo approaches to study angiogenesis in the pathophysiology and therapy of endometriosis Human Reproduction Update 2007, 13 (4):331 342. 79. Dewhirst M: Concepts of oxygen transport at the microcirculatory level Semin Radiat Oncol 1998, 8 :143 150. 80. Jain RK, Munn LL, Fukumura D: Dissecting tumour pathophysiology using intravital microscopy Nat Rev Cancer 2002, 2 (4):266 276. 81. Leunig M, Yuan F, Menger MD, Boucher Y, Goetz AE, Messmer K, Jain RK: Angiogenesis, Microvascular Ar chitecture, Microhemodynamics, and Interstitial Fluid Pressure During Early Growth of Human Adenocarcinoma Ls174t in Scid Mice Cancer Research 1992, 52 (23):6553 6560. 82. Li CY, Shan SQ, Cao YT, Dewhirst MW: Role of incipient angiogenesis in cancer metas tasis Cancer and Metastasis Reviews 2000, 19 (1 2):7 11. 83. Abdul Karim MA, Al Kofahi K, Brown EB, Jain RK, Roysam B: Automated tracing and change analysis of angiogenic vasculature from in vivo multiphoton confocal image time series Microvasc Res 2003, 66 (2):113 125. 84. Friedl P, Wolf K: Proteolytic and non proteolytic migration of tumour cells and leucocytes In: Proteases and the Regulation of Biological Processes. Edited by Saklatvala J, Nagase H, Salvesen G, vol. 70. London: Portland Press Ltd; 20 03: 277 285. 85. Kashiwagi S, Izumi Y, Gohongi T, Demou ZN, Xu L, Huang PL, Buerk DG, Munn LL, Jain RK, Fukumura D: NO mediates mural cell recruitment and vessel morphogenesis in murine melanomas and tissue engineered blood vessels Journal of Clinical In vestigation 2005, 115 (7):1816 1827. 86. Garini Y, Young IT, McNamara G: Spectral imaging: Principles and applications Cytom Part A 2006, 69A (8):735 747.
113 87. Wyatt CL: Infrared Spectrometer Liquid Helium Cooled Rocketborne Circular Variable Filter App lied Optics 1975, 14 (12):3086 3092. 88. Miller PJ: Use of Tunable Liquid Crystal Filters to Link Radiometric and Photometric Standards Metrologia 1991, 28 (3):145 149. 89. Shonat RD, Wachman ES, Niu W, Koretsky AP, Farkas DL: Near simultaneous hemoglobin saturation and oxygenation tension maps in mouse brain using an AOTF microscope Biophysical Journal 1997, 73 :1223 1231. 90. Sorg BS, Moeller BJ, Donovan O, Cao Y, Dewhirst MW: Hyperspectral imaging of hemoglobin saturation in tumor microvasculature and tumor hypoxia development J Biomed Opt 2005, 10 (4): Research Papers. 91. Perry BN, Arbiser JL: The duality of angiogenesis: implications for therapy of human disease Journal of Investigative Dermatology 2006, 126 :2160 2166. 92. Otrock ZK, Mahfouz RAR, Makarem JA, Shamseddine AI: Understanding the biology of angiogenesis: Review of the most important molecular mechanisms Blood Cells, Molecules, and Diseases 2007, 39 :212 220. 93. Carmeliet P, Jain RK: Angiogenesis in cancer and other diseases In: Natur e. vol. 407; 2000: 249 257. 94. Dewhirst MW: Angiogenesis and blood flow in solid tumors In: Drug Resistance in Oncology. Edited by Teicher BA. New York: Marcel Dekker; 1993: 3 23. 95. Warren BA: The vascular morphology of tumors In: Tumor blood circul ation: Angiogenesis, vascular morphology and blood flow of experimental and human tumors. Edited by Peterson HI. Boca Raton: CRC Press, Inc.; 1979: 1 47. 96. Simons M: Angiogenesis, arteriogenesis, and diabetes paradigm reassessed? Journal of the Americ an College of Cardiology 2005, 46 (5):835 837. 97. Martin A, Komada MR, Sane DC: Abnormal angiogenesis in diabetes mellitus Medicinal Research Reviews 2003, 23 (2):117 145. 98. Carvalho JF, Blank M, Shoenfeld Y: Vascular endothelial growth factor (VEGF) i n autoimmune diseases Journal of Clinical Immunology 2007, 27 (3):246 256.
114 99. Bainbridge J, Sivakumar B, Paleolog E: Angiogenesis as a therapeutic target in arthritis: lessons from oncology Current Pharmaceutical Design 2006, 12 :2631 2644. 100. Azfar R S, Gelfand JM: Psoriasis and metabolic disease: epidemiology and pathophysiology Current Opinion in Rheumatology 2008, 20 :416 422. 101. Gargiulo P, Giusti C, Pietrobono D, La Torre D, Diacono D, Tamburrano G: Diabetes mellitus and retinopathy Digestive a nd Liver Disease 2004, 36 (Suppl 1) :S101 S105. 102. Wegewitz U, Ghring I, Spranger J: Novel approaches in the treatment of angiogenic eye disease Current Pharmaceutical Design 2005, 11 :2311 2330. 103. Safety of sodium fluorescein f or in vivo study of skin Skin Research and Technology 2006, 12 :155 161. 104. Werkmeister E, Kerdjoudj H, Marchal L, Stoltz JF, Dumas D: Multiphoton microscopy for blood vessel imaging: new non invasive tools (Spectral, SHG, FLIM) Clinical Hemorheology a nd Microcirculation 2007, 37 :77 88. 105. Schaffer CB, Friedman B, Nishimura N, Schroeder LF, Tsai PS, Ebner FF, Lyden PD, Kleinfeld D: Two photon imaging of cortical surface microvessels reveals a robust redistribution in blood flow after vascular occlusi on PLoS Biology 2006, 4 :e22. 106. Reyes Aldasoro CC, Wilson I, Prise VE, Barber PR, Ameer Beg M, Vojnovic B, Cunningham VJ, Tozer GM: Estimation of apparent tumor vascular permeability from multiphoton fluorescence microscopic images of P22 rat sarcomas in vivo Microcirculation 2008, 15 (1):65 79. 107. Fu D, Matthews TE, Ye T, Piletic IR, Warren WS: Label free in vivo optical imaging of microvasculature and oxygenation level Journal of Biomedical Optics 2008, 13 (4):article 040503. 108. Choi B, Kang NM, Nelson JS: Laser speckle imaging for monitoring blood flow dynamics in the in vivo rodent dorsal skin fold model Microvasc Res 2004, 68 :143 146. 109. Dunn AK, Devor A, Bolay H, Andermann ML, Moskowitz MA, Dale AM, Boas DA: Simultaneous imaging of total cerebral hemoglobin concentration, oxygenation, and blood flow during functional activation Optics Letters 2003, 28 (1):28 30. 110. Kolkman RGM, Klaessens JHGM, Hondebrink E, Hopman JCW, de Mul FFM, Steenbergen W, Thijssen JM, van Leeuwen TG: Photoacousti c determination
115 of blood vessel diameter Physics in Medicine and Biology 2004, 49 :4745 4756. 111. Wang X, Ku G, Wegiel MA, Bornhop DJ, Stoica G, Wang LV: Noninvasive photoacoustic angiography of animal brains in vivo with near infrared light and an optic al contrast agent Optics Letters 2004, 29 (7):730 732. 112. Wang X, Xie X, Ku G, Wang LV, Stoica G: Noninvasive imaging of hemoglobin concentration and oxygenation in the rat brain using high resolution photoacoustic tomography Journal of Biomedical Opti cs 2006, 11 (2):article 024015. 113. Hitzenberger C, Trost P, Lo PW, Q. Zhou, Opt. Express 11, 2753 2761 (2003): Three dimensional imaging of the human retina by high speed optical coherence tomography Optics Express 2003, 11 :2753 2761. 114. Izatt JA, Kulkarni MD, Yazdanfar S, Barton JK, Welch AJ: In vivo bidirectional color Doppler flow imaging of picoliter blood volumes using optical coherence tomography Optics Letters 1997, 22 :1439 1441. 115. Michaely R, Bachmann AH, Villiger ML, Blatter C, Lasser T, Leitgeb RA: Vectorial reconstruction of retinal blood flow in three dimensions measured with high resolution resonant Doppler Fourier domain optical coherence tomography Journal of Biomedical Optics 2007, 12 (4):article 041213. 116. Faber DJ, Mick EG, Aalders MCG, van Leeuwen TG: Toward assessment of blood oxygen saturation by spectroscopic optical coherence tomography Optics Letters 2005, 30 :1015 1017. 117. Faber DJ, Mik EG, Aalders MCG, van Leeuwen TG: Light absorption of (oxy )hemoglobin assessed b y spectroscopic optical coherence tomography Optics Letters 2003, 28 (16):1436 1438. 118. Lu CW, Lee CK, Tsai MT, Wang YM, Yang CC: Measurement of the hemoglobin oxygen saturation level with spectroscopic spectral domain optical coherence tomography Opti cs Letters 2008, 33 (5):416 418. 119. Kagemann L, Wollstein G, Wojtkowski M, Ishikawa H, Townsend KA, Gabriele ML, Srinivasan VJ, Fujimoto JG, Schuman JS: Spectral oximetry assessed with high speed ultra high resolution optical coherence tomography Journa l of Biomedical Optics 2007, 12 (4):article 041212. 120. Iftimia NV, Hammer DX, Bigelow CE, Rosen DI, Ustun T, Ferrante AA, Vu D, Ferguson RD: Toward noninvasive measurement of blood hematocrit using
116 spectral domain low coherence interferometry and retinal tracking Optics Express 2006, 14 :3377 3388. 121. Mariampillai A, Standish BA, Moriyama EH, Khurana M, Munce NR, Leung MKK, Jiang J, Cable A, Wilson BC, Vitkin A et al : Speckle variance detection of microvasculature using swept source optical coherence t omography Optics Letters 2008, 33 (13):1530 1532. 122. Styp Rekowska B, Disassa NM, Reglin B, Ulm L, Kuppe H, Secomb TW, Pries AR: An imaging spectroscopy approach for measurement of oxygen saturation and hematocrit during intravital microscopy Microcirc ulation 2007, 14 :207 221. 123. Kobayashi H, Takizawa N: Imaging of oxygen transfer among microvessels of rat cremaster muscle Circulation 2002, 105 :1713 1719. 124. Sorg BS, Hardee ME, Agarwal N, Moeller BJ, Dewhirst MW: Spectral imaging facilitates visu alization and measurements of unstable and abnormal microvascular oxygen transport in tumors Journal of Biomedical Optics 2008, 13 (1):article 014026. 125. Sorg BS, Moeller BJ, Donovan O, Cao Y, Dewhirst MW: Hyperspectral imaging of hemoglobin saturation in tumor microvasculature and tumor hypoxia development Journal of Biomedical Optics 2005, 10 :044004. 126. Skala MC, Fontanella A, Hendargo H, Dewhirst MW, Izatt JA: Combined hyperspectral and spectral domain optical coherence tomography microscope for n oninvasive hemodynamic imaging Optics Letters 2009, 34 (3):289 291. 127. Shonat RD, Johnson PC: Oxygen tension gradients and heterogeneity in venous microcirculation: a phosphorescence quenching study American Journal of Physiology Heart and Circulator y Physiology 1997, 272 :H2233 H2240. 128. Unthank JL, Lash JM, Nixon JC, Sidner RA, Bohlen HG: Evaluation of carbocyanine labeled erythrocytes for microvascular measurements Microvasc Res 1993, 45 :193 210. 129. Shonat RD, Wachman ES, Niu WH, Koretsky AP, Farkas DL: Near simultaneous hemoglobin saturation and oxygen tension maps in mouse brain using an AOTF microscope Biophysical Journal 1997, 73 :1223 1231. 130. Ferreira AJ, Raizada MK: Genomic and proteomic approaches for targeting of angiotensin conver ting enzyme2 for cardiovascular diseases Current Opinion in Cardiology 2008, 23 (4):364 369.
117 131. Santos RA, Ferreira AJ, Simes E, Silva AC: Recent advances in the angiotensin converting enzyme 2 angiotensin(1 7) Mas axis Experimental Physiology 2008, 9 3 (5):519 527. 132. Katayama T, Ikeda Y, Handa M, Tamatani T, Sakamoto S, Ito M, Ishimura Y, Suematsu M: Immunoneutralization of glycoprotein Ib attenuates endotoxin induced interactions of platelets and leukocytes with rat venular endothelium in vivo Ci rculation Research 2000, 86 (10):1031 1037. 133. Park SO, Lee YJ, Seki T, Hong KH, Fliess N, Jiang Z, Park A, Wu X, Kaartinen V, Roman BL et al : ALK5 and TGFBR2 independent role of ALK1 in the pathogenesis of hereditary hemorrhagic telangiectasia type 2 Blood 2008, 11 (2):633 642. 134. Vaupel P, Kallinowski F, Okunieff P: Blood flow, oxygen and nutrient supply, and metabolic microenvironment of human tumors: a review Cancer Res 1989, 49 :6449 6465. 135. Bristow RG, Hill RP: Hypoxia, DNA repair and geneti c instability Nature Reviews Cancer 2008, 8 :180 192. 136. Durand RE, Aquino Parsons C: Clinical relevance of intermittent tumour blood flow Acta Oncologica 2001, 40 (8):929 936. 137. Durand RE, Aquino Parsons C: Non constant tumour blood flow implic ations for therapy Acta Oncologica 2001, 40 (7):862 869. 138. Rofstad EK, Galappathi K, Mathiesen B, Ruud E BM: Fluctuating and diffusion limited hypoxia in hypoxia induced metastasis Clinical Cancer Research 2007, 13 :1971 1978. 139. Brurberg KG, Gaustad JV, Mollatt CS, Rofstad EK: Temporal heterogeneity in blood supply in human tumor xenografts Neoplasia 2008, 10 (7):727 735. 140. Sorg BS, Hardee ME, Agarwal N, Moeller BJ, Dewhirst MW: Spectral imaging facilitates visualization and measurements of unstable and abnormal microvascular oxygen transport in tumors Journal of Biomedical Optics 2008, 13 (1). 141. Gaustad JV, Simonsen TG, Brurberg KG, Huuse EM, Rofstad EK: Blood Supply in Melanoma Xenografts Is Governed by the Morpholog y of the Supplying Arteries Neoplasia 2009, 11 (3):277 285.
118 142. Hardee ME, Dewhirst MW, Agarwal N, Sorg BS: Novel imaging provides new insights into mechanisms of oxygen transport in tumors Current Molecular Medicine 2008:accepted. 143. Fitzgerald DJ: Vascular biology of thrombosis: The role of platelet vessel wall adhesion Neurology 2001, 57(Suppl 2) :S1 S4. 144. Kyrle PA, Eichinger S: Deep vein thrombosis Lancet 2005, 365 :1163 1174. 145. Fraga Silva RA, Ferreira AJ, Sorg BS, Wankhede M, deDeugd C, Jun JY, Baker MB, Castellano RJ, Katovich MJ, Raizada MK: Pharmacological ACE2 activation exerts antithrombotic activity In: American Heart Association International Stroke Conference. San Diego, CA; 2009. 146. Fraga Silva R, Sorg BS, Wankhede M, DeDeugd C, Jun J, Baker M, Li Y, Castello R, Katovich M, Raizada MK et al : ACE2 Activation Promotes Anti Thrombotic Activity Molecular Medicine 2010, In press 147. Konner K: The initial creation of native arteriovenous fistulas: surgical aspects and their impa ct on the practice of nephrology Seminars in Dialysis 2003, 16 (4):291 298. 148. Jain RK: Determinents of tumor blood flow: A review Cancer Research 1988, 48 :2641 2658. 149. Widdicombe J: Microvascular anatomy of the nose Allergy 1997, 52 (40):(Suppl) 7 11. 150. Zakrzewicz A, Secomb TW, Pries AR: Angioadaptation: Keeping the vascular system in shape News in Physiological Sciences 2002, 17 :197 201. 151. Guyton AC: Textbook of Medical Physiology 8th edn. Philadelphia, PA: W. B. Saunders Company; 1991. 152. Park SO, Wankhede M, Lee YJ, Choi EJ, Fliess N, Choe S, Oh SH, Walter G, Raizada MK, Sorg BS et al : Real time imaging of de novo arteriovenous malformation in a mouse model of hereditary hemorrhagic telangiectasia J Clin Invest 2009, 119 :3487 3496. 153. Denekamp J: Endothelial cell proliferation as a novel approach to targeting tumour therapy Br J Cancer 1982, 45 :136 139 154. Denekamp J, Hill SA, Hobson B: Vascular occlusion and tumour cell death Eur J Cancer Clin Oncl 1983, 19 :271 275.
119 155. C haplin DJ, Pettit GR, Hill SA: Anti vascular approaches to solid tumour therapy: evaluation of combretastatin A4 phosphate Anticancer Res 1999, 19 :189 195. 156. Siemann DW, Chaplin DJ, Walicke PA: A review and update of the current status of the vasculat ure disabling agent combretastatin A4 phosphate (CA4P) Expert Opinion on Investigational Drugs 2009, 18 :189 197. 157. Siemann DW, Horsman MR: Vascular targeted therapies in oncology Cell and Tissue Research 2009, 335 :241 248. 158. Salmon HW, Mladinich C, Siemann DW: Evaluations of vascular disrupting agents CA4P and OXi4503 in renal cell carcinoma (Caki 1) using a silicon based microvascular casting technique E uropean Journal of Cancer 2006, 42 :3073 3078. 159. Kimura H, Braun RD, Ong ET, Hsu R, Secom b TW, Papahadjopoulos D, Hong K, Dewhirst MW: Fluctuations in red cell flux in tumor microvessels can lead to transient hypoxia and reoxygenation in tumor parenchyma Cancer Research 1996, 56 :5522 5528. 160. Sorg BS, Moeller BJ, Donovan O, Cao Y, Dewhirst MW: Measurement of hemoglobin saturation in tumor microvasculature in preclinical models using hyperspectral imaging Proceedings of SPIE 2005, 5694 :74 81. 161. Jordan A, Hadfield JA, Lawrence NJ, McGown AT: Tubulin as a Target for Anticancer Drugs:Agent s Which Interact with the Mitotic Spindle Med Res Rev 1998, 18 (4):259 296. 162. Lin CM, Singh SB, Chu PS, Dempcy R, Schmidt JM, Pettit GR, Hamel E: Interactions of Tubulin with Potent Natural and Synthetic Analogs of the Antimitotic Agent Combretastatin: a Structure Activity Study M olecular Pharmacology 1988, 34 :200 208. 163. Folkes LK, Christlieb M, Madej E, Stratford MRL, Wardman P: Oxidative Metabolism of Combretastatin A 1 Produces Quinone Intermediates with the Potential To Bind to Nucleophiles and To Enhance Oxidative Stress via Free Radicals Chem Res Toxicol 2007, 20 :1885 1894. 164. Salmon BA, Siemann DW: Characterizing the tumor response to treatment with combretastatin A4 phosphate Int J Radiation Oncology Biol Phys 2007, 68 (1):211 217. 165 Tozer GM, Ameer Beg SM, Baker J, Barber PR, Hill SA, Hodgkiss RJ, Locke R, Prise VE, Wilson I, Vojnovic B: Intravital imaging of tumour vascular
120 networks using multi photon fluorescence microscopy Advanced Drug Delivery Reviews 2005, 57 :135 152. 166. S iemann DW, Rojiani AM: T he Vascular disrupting Agent ZD6126 shows increased antitumor efficacy and enhanced radiation response in large advanced tumors Int J Radiation Oncology Biol Phys 2005, 62 (3):846 853. 167. Parkins CS, Holder AL, Hill SA, Chaplin D J, Tozer GM: Determinants of anti vascular action by combretastatin A 4 phosphate: role of nitric oxide British Journal of Cancer 2000, 83 (6):811 816 168. Skliarenko JV, Lunt SJ, Gordon ML, Vitkin A, Milosevic M, Hill RP: Effects of the Vascular Disrupt ing Agent ZD6126 on Interstitial Fluid Pressure and Cell Survival in Tumors Cancer Res 2006, 66 (4):2074 2080. 169. Dalal S, Burchill SA: Preclinical evaluation of vascular disrupting agents in European Journal of Cancer 2009, 45 :713 722. 170. Folkman J: How is blood vessel growth regulated in normal and neoplastic tissue? Cancer Res 1986, 46 :467 473. 171. Landuyt W, Verdoes O, Darius D, Drijkoningen M, Nuyts S, Theys J, Stockx L, Wynendaele W, Fowler J, Maleux G et al : Vascular targeting of solid tumours: a major `inverse' volume response relationship following combretastatin A 4 phosphate treatment of rat rhabdomyosarcomas European Journal of Cancer 2000, 36 :1833 1843. 172. Masunaga S I Nagasawa H, Nagata K, Suzuki M Uto Y, Hori H, Kinashi Y, Ono K: Dependency of the effect of a vascular disrupting agent on sensitivity to tirapazamine and gamma ray irradiation upon the timing of its administration and tumor size, with reference to the effect on intratumor quiescent c ells J Cancer Res Clin Oncol 2007, 133 :47 55. 173. Pettit GR, Singh SB, Hamel E, Lin C, Alberts D, Garcia Kendall: Isolation and structure of the strong cell growth and tubulin inhibitor combretastatin A 4 Experientia 1989, 45 :209 211. 174. Wankhede M, DeDeugd C, Siemann DW, Sorg BS: In vivo functional differences in microvascular response of 4T1 and Caki 1 tumors after treatment with OXi4503 Oncology Reports 2010, 23 :685 692. 175. Sharathkumar, A.A. & Shapiro, A. Hereditary haemorrhagic telangiectasi a. Haemophilia 2008, 14 : 1269 1280.
121 176. Berg, J. et al. Hereditary hemorrhagic telangiectasia: A questionnaire based study to delineate the different phenotypes caused by endoglin and ALK1 mutations J Med Genet 2003 40 : 585 590. 177. Sadick, H. et al. Hereditary hemorrhagic telangiectasia: An update on clinical manifestations and diagnostic measures Wien. Klin. Wochenschr 2006 118 :72 80. 178. Willoughby, S., Holmes, A. & loscalzo, J. Platelets and cardiovascular disease Eur J Cardiovasc Nurs 2002 1 : 273 288. 179. Braverman, I.M., Keh, A. & Jacobson, B.S. Ultrastructure and three dimensional organization of the telangiectases of hereditary hemorrhagic telangiectasia J Invest Dermatol 1990 95 :422 427. 180. Murphy, P.A. et al. Endothelial Notch4 sign alling induces hallmarks of brain arteriovenous malformations in mice Proc N atl Acad Sci 2008 105 :10901 10906.
122 BIOGRAPHICAL SKETCH Mamta Wankhede obtain ed her b Nagpur University, India ( 1999 2003) and joined the Biomedical Engineering program at U niversity of F lorida for a Master of Science degree (2005 2008 ). A fter completion of her PhD program in biomedical e ngineering in Dec 2010 she aspires to remain in the research field in future.