DEVELOPMENT AND CHARACTERIZATION OF A ... - TSpace

DEVELOPMENT AND CHARACTERIZATION OF A LIPOSOME IMAGING AGENT

by

Jinzi Zheng

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Graduate Department of Medical Biophysics University of Toronto

© Copyright by Jinzi Zheng (2009)

Abstract Development and Characterization of a Liposome Imaging Agent Doctor of Philosophy, 2009 Jinzi Zheng Department of Medical Biophysics University of Toronto

Applied cancer research is heavily focused on the development of diagnostic tools with high sensitivity and specificity that are able to accurately detect the presence and anatomical location of neoplastic cells, as well as therapeutic strategies that are effective at curing or controlling the disease while being minimally invasive and having negligible side effects. Recently, much effort has been placed on the development of nanoparticles as diagnostic imaging and therapeutic agents, and several of these nanoplatforms have been successfully adopted in both the research and clinical arenas. This thesis describes the development of a nanoparticulate liposome system for use in a number of applications including multimodality imaging with computed tomography (CT) and magnetic resonance (MR), longitudinal vascular imaging, imagebased biodistribution assessment, and CT detection of neoplastic and inflammatory lesions. Extensive in vitro and in vivo characterization was performed to determine the physico-chemical properties of the liposome agent, including its size, morphology, stability and agent loading, as well as its pharmacokinetics, biodistribution, tumor targeting and imaging performance. Emphasis was placed on the in vivo CT-based quantification of liposome accumulation and clearance from healthy and tumor tissues in

ii

a VX2 carcinoma rabbit model, gaining insight not only on the spatial but also the temporal biodistribution of the agent. The thesis concludes with a report that describes the performance of liposomes and CT imaging to detect and localize tumor and inflammatory lesions as compared to that of

18

F-fluorodeoxyglucose (FDG) – positron

emission tomography (PET). The outcome of the study suggests that liposome-CT could be employed as a competitive method for whole body image-based disease detection and localization. Overall, this work demonstrated that this liposome agent along with quantitative imaging systems and analysis tools, has the potential to positively impact cancer treatment outcome through improved diagnosis and staging, as well as enable personalization of treatment delivery via target delineation. However, in order to prove clinical benefit, steps must be taken to advance this agent through the regulatory stages and obtain approval for its use in humans. Ultimately, the clinical adoption of this multifunctional agent may offer improvements for disease detection, spatial delineation and therapy guidance.

iii

Acknowledgements

The successful completion of this thesis was made possible by various contributions from a number of people. Their scientific contributions are acknowledged at the end of each pertinent chapters of this thesis. Here I would like to recognize and thank individuals who have contributed to my growth both as a person and as a scientist over the course of the past five years:

-

My supervisors Dr. David Jaffray and Dr. Christine Allen for their ongoing support and encouragement, in addition to their scientific guidance and mentorship;

-

My supervisory committee members Dr. Mark Henkelman, Dr. Sandy Pang and Dr. Cynthia Menard for providing me with a fresh outlook on research problems and help in finding potential solutions;

-

Fellow labmates, past and present, for their company during many many lunches and coffee breaks, for their constructive criticism during all of my practice talks, for the fun times at conferences, and above all for their sincere and generous friendship;

-

My boyfriend Patrick Blit for his patience during my late night experiments, his willingness to provide feedback on many conference abstracts and scholarship applications, as well as for his constant love and support throughout the highs and lows of my graduate school journey;

-

My parents for never doubting my capabilities and always pushing me to perform at my very best.

iv

Table of Contents Chapter 1.

Introduction................................................................................................. 1

1.1.

Nanoparticles in Cancer Diagnosis and Treatment............................................. 2

1.2.

Rationale for Spatio-Temporal Biodistribution Assessment .............................. 6

1.3.

Imaging as a Non-invasive Method for Nanoparticle Biodistribution

Assessment...................................................................................................................... 8 1.4.

Thesis Outline ................................................................................................... 11

Chapter 2.

Multimodal Contrast Agent for Combined CT and MR Imaging

Applications ................................................................................................................... 14 2.1.

Foreword ........................................................................................................... 15

2.2.

Introduction....................................................................................................... 15

2.3.

Materials and Methods...................................................................................... 19

2.4.

Results............................................................................................................... 23

2.5.

Discussion ......................................................................................................... 36

2.6.

Acknowledgements........................................................................................... 40

Chapter 3.

In Vivo Performance of a Liposomal Vascular Contrast Agent for CT and

MR-Based Image Guidance Applications ........................................................................ 41 3.1.

Foreword ........................................................................................................... 42

3.2.

Introduction....................................................................................................... 42

v

3.3.

Materials and Methods...................................................................................... 44

3.4.

Results............................................................................................................... 50

3.5.

Discussion ......................................................................................................... 61

3.6.

Acknowledgements........................................................................................... 64

Chapter 4.

Quantitative CT Imaging of the Spatial and Temporal Distribution of

Liposomes in a Rabbit Tumor Model ............................................................................... 65 4.1.

Foreword ........................................................................................................... 66

4.2.

Introduction....................................................................................................... 66

4.3.

Experimental Section ........................................................................................ 68

4.4.

Results............................................................................................................... 71

4.5.

Discussion ......................................................................................................... 84

4.6.

Acknowledgements........................................................................................... 88

Chapter 5.

Liposome Contrast Agent for CT-based Detection and Localization of

Neoplastic and Inflammatory Lesions in Rabbits: Validation with FDG-PET and Histology

................................................................................................................... 89

5.1.

Foreword ........................................................................................................... 90

5.2.

Introduction....................................................................................................... 90

5.3.

Materials and Methods...................................................................................... 92

5.4.

Results............................................................................................................... 98

5.5.

Discussion ....................................................................................................... 109 vi

5.6.

Acknowledgements......................................................................................... 112

Chapter 6.

Summary and Future Directions ............................................................. 113

6.1.

Summary ......................................................................................................... 114

6.2.

Future Directions ............................................................................................ 115

6.2.1.

Technology

Translation

and

Commercialization:

Challenges

and

Opportunities........................................................................................................... 116 6.2.2.

Extension to a Modular Multimodality Imaging Platform ..................... 117

6.2.3.

Additional Characterization of Liposome Transport and Distribution ... 121

References....................................................................................................................... 125

vii

List of Tables Table 2.1

Size and loading characteristics of the dual-agent-containing liposome

formulation................................................................................................................ 25 Table 2.2

Relaxivity r1 and r2 values for the free gadoteridol, free iohexol and

gadoteridol, free iohexol and liposome encapsulated agents solutions. ................... 33 Table 3.1 Pharmacokinetic parameters for iohexol and gadoteridol when administered in a liposome formulation to female Balb-C mice. . ..................................................... 53 Table 4.1 Liposome biodistribution expressed as %ID and as %ID/cm3 of organ/tissue. . ................................................................................................................................... 78 Table 4.2 List of the iodine concentration detection sensitivity using CT for organ and tissues of known volumes. . ...................................................................................... 80 Table 5.1 List and classification of the neoplastic and inflammatory lesions detected using CT and PET imaging, their volumes and maximum signal values. . .............. 99

viii

List of Figures Figure 2.1 Schematic of the liposome-based contrast agent system................................ 18 Figure 2.2

Transmission electron micrograph of the negatively stained dual-agent

containing liposomes. ............................................................................................... 24 Figure 2.3 The in vitro release profile for iohexol and gadoteridol from liposomes dialyzed under sink conditions against HBS at 4 °C and 37 °C ............................... 26 Figure 2.4

Size of the dual-agent-containing liposomes during dialysis under sink

conditions against HBS at 37 °C............................................................................... 27 Figure 2.5 In vitro imaging efficacy of the liposome-based contrast agent system in CT and MR...................................................................................................................... 28 Figure 2.6

CT and MR signals as a function of increasing iodine and gadolinium

concentrations. .......................................................................................................... 30 Figure 2.7

1/T1 and 1/T2 relaxation rates as a function of gadolinium and iodine

concentrations. .......................................................................................................... 32 Figure 2.8 Illustration of the use of the liposome-based contrast agent in a healthy rabbit model in CT and MR. ............................................................................................... 34 Figure 2.9 Relative percentage signal enhancement achieved in the aorta of the rabbit measured from MR and CT images. ......................................................................... 35 Figure 3.1 Pharmacokinetics of free iohexol, free gadoteridol, liposomal iohexol and liposomal gadoteridol in healthy female Balb-C mice.. ........................................... 52 Figure 3.2 Biodistribution of iohexol and gadoteridol when administered in a liposome formulation to female Balb-C mice. ........................................................................ 55

ix

Figure 3.3 CT and MR 3D maximum intensity projection images of a healthy New Zealand White rabbit pre and post liposome administration. ................................... 56 Figure 3.4 Plots of the relative change in signal intensity pre- and post-administration of the multimodal liposomal agent in CT and MR versus the measured plasma iodine and gadolinium concentrations.. ............................................................................... 59 Figure 3.5 Summary of the hematological and biochemical evaluation of plasma samples obtained from female Balb-C mice........................................................................... 60 Figure 4.1 Liposome biodistribution and kinetics in tumor-bearing rabbits assessed via CT imaging. .............................................................................................................. 73 Figure 4.2 (a) Liposome biodistribution profiles in the various organs and tissues of interest as measured using CT-based detection of the co-encapsulated iohexol and gadoteridol. (b) Time-dependent tumor-to-muscle ratio of iodine concentration. ... 75 Figure 4.3 CT maximum intensity projections of a representative tumor-bearing rabbit and of five segmented tumor volumes pre and post liposome injection................... 81 Figure 4.4 Tumor volume fraction occupied by liposomes and the change in tumor volumes measured using CT in the five rabbits over 14 days. ................................. 83 Figure 5.1 Flow chart illustration of the experimental steps. .......................................... 97 Figure 5.2 Three cases of primary tumors detected by CT and PET, and confirmed by histology.................................................................................................................. 103 Figure 5.3 Two cases of inflammatory lesions in the muscle detected by CT and PET, and confirmed by histology.. .................................................................................. 104 Figure 5.4

CT and PET imaging signal intensities of neoplastic and inflammatory

lesions. .................................................................................................................... 106

x

Figure 5.5

Kinetic profiles of liposome contrast agent accumulation and clearance in

tumor and inflammatory lesions. ............................................................................ 107 Figure 5.6 Incidental finding: malignant lymph nodes detected by FDG-PET ............ 108 Figure 6.1 Schematic representation of the modular multimodality liposome imaging platform................................................................................................................... 120

xi

List of Abbreviations and Symbols

%ID

Percent injected dose

ALP

Alkaline phosphatase

ALT

Alanine transaminase

AST

Aspartate transaminase

AUC

Area under the curve

CH

Cholesterol

CIHR

Canadian Institute of Health Research

CL

Plasma clearance

CT

Computed tomography

DCE-MR

Dynamic contrast enhanced-MR

∆HU

Change in Hounsfield unit

DLS

Dynamic light scattering

∆meanHU

Change in mean Hounsfield unit

DNA

Deoxyribonucleic acid

DPPC

1,2-Dipalmitoyl-sn-Glycero-3-Phosphocholine

DSPE

1,2-Distearoyl-sn-Glycero-3-Phosphoethanolamine

DTPA

Diethylene triamine pentaacetic acid

EPR

Enhanced permeability and retention

FAZA

Fluoroazomycin arabinoside

FDG

Fluoro-2-deoxy-D-glucose

FMISO

Fluoromisonidazole

xii

FOV

Field of view

GMP

Good manufacturing practice

H&E

Hematoxylin and eosin

HBS

HEPES buffer solution

HBSS

Hanks balanced salt solution

HEPES

4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid

HPLC

High performance liquid chromatography

HSPC

Hydrogenated soya phosphatidylcholine

HU

Hounsfield unit

ICP-AES

Inductively coupled plasma atomic emission spectrometry

IR

Inversion recovery

Kd

Distribution rate constant

Ke

Elimination rate constant

MHC

Major histocompatibility complex

MIP

Maximum intensity projection

MPS

Monophagocytic system

MR

Magnetic resonance

MRS

Magnetic resonance spectroscopy

MWCO

Molecular weight cut-off

PA

Phosphatidic acid

pan-CK

Pan-cytokeratin

PBS

Phosphate buffered saline

PEG

Poly-[ethylene glycol]

xiii

PET

Positron emission tomography

PK

Pharmacokinetics

PS

Phosphatidylserine

PTA

Phosphotungstic acid

R2

Coefficient of determination

r1

Longitudinal relaxivity

r2

Transverse relaxivity

RBC

Red blood cells

RES

Reticulo-endothelial system

ROI

Region(s) of interest

SI

Signal intensity

SPECT

Single photon emission computed tomography

SUVmax

Maximum standardized uptake value

t1/2

Half-life (vascular or physical)

t1/2α

Distribution half-life

t1/2β

Elimination half-life

TE

Echo time

TEM

Transmission electron microscopy

T1

Longitudinal relaxation rate constant

T2

Transverse relaxation rate constant

TI

Inversion time

TLD

Thermo luminescent dosimeter

TR

Repetition time

xiv

UHN

University Health Network

USPIO

Ultrasmall superparamagnetic iron oxide

UV

Ultraviolet

Vd

Volume of distribution

WBC

White blood cells

xv

Chapter 1. Introduction

1

2 This thesis reports on the development of a liposomal nanoparticle system that supports non-invasive characterization of its in vivo biodistribution and kinetics by volumetric imaging modalities such as computed tomography (CT) and magnetic resonance (MR) imaging. Furthermore, its potential applications in cancer diagnosis and treatment are also explored.

Specifically, proof-of-principle studies were conducted to assess the

feasibility of its employment for multimodality imaging, blood pool/vascular imaging, image-guided assessment of drug delivery, and cancer detection. The following chapter outlines the rationale as well as provides literature background to frame the context of the work described in this thesis.

1.1.

Nanoparticles in Cancer Diagnosis and Treatment The two main goals in applied cancer research are the development of 1) diagnostic

tools with high sensitivity and specificity that are able to accurately detect the presence and anatomical location of neoplastic cells as well as characterize their abnormal nature; and 2) therapeutic strategies that are effective at curing or controlling the disease, while being minimally invasive and having negligible side effects. Recently, much effort has been placed on the development of nanoparticles as diagnostic imaging and therapeutic agents, and several of these nanoplatforms have achieved success in both the research and clinical arenas. Overall, there are three rationales for employing nanoparticles instead of traditional small molecules as a new class of agents that have the potential to lead to improved diagnosis and treatment. First, the critical size of nanoparticles and their tunable surface characteristics result in pharmacokinetics profiles that enable applications requiring longitudinal imaging and sustained drug delivery. Second, their extended pharmacokinetics allow for increased

3 tumor tissue targeting resulting in greater target-to-background signal ratio in imaging applications and enhanced therapeutic index when used as a treatment vector. Third, the high payload of imaging and/or therapeutic agents that nanoparticles carry can be exploited for amplification of imaging signal or therapeutic effect, especially when used in conjunction with a lower sensitivity imaging system or a less cytotoxic drug, respectively. The pharmacokinetics of a nanoparticle is determined by its size, charge, surface modification and shape [1]. It is generally agreed that the hydrodynamic radius of a nanoparticles should be at least 10 nm [2] (greater than the sieving coefficient of the renal glomerular capillary wall for spherical particles) in order to avoid significant clearance via renal excretion [3, 4] as well as distribution into the extracellular space through the fenestrations in the vascular endothelial walls (up to 10 nm [5]) . In addition to the above described passive size-dependent nanoparticle removal process, the body also has an active nanoparticle opsonization process in place by the reticulo-endothelial system (RES) [6] also termed mononuclear phagocytic system (MPS). The most commonly used strategy to minimize opsonization is particle surface modification with poly-[ethylene glycol] (PEG) [7]. PEG is an inert and highly hydrophilic linear polymer chain. Its incorporation onto the surface of nanoparticles provides good steric stabilization and prevents both self-aggregation as well as protein binding [1]. As a result, nanoparticle destabilization due to protein adhesion is minimized and its vascular circulation lifetime is increased. Sadzuka et al. investigated the pharmacokinetics of drugs encapsulated in either non-PEGylated or PEGylated liposomes. They observed a 6-fold increase in the area under the curve (AUC) of the drug pharmacokinetics profile when it is formulated in the PEGylated liposomes compared to the PEG-free liposome formulation and a 36-fold increase compared to the

4 AUC of the free drug [8]. The size of nanoparticles also plays an important role in their accumulation in organs that make up the RES, namely liver and spleen. Liu et al. reported in 1992 on the biodistribution of liposomes of different sizes (30 – 450 nm) [9]. They measured 70% of the injected dose (%ID) of liposomes with a diameter less than 70 nm localized in the liver. This can be explained by the size of the fenestrations in the endothelium of the liver sinusoid (100 nm, [10]). In the spleen, Liu et al. found that liposomes with a diameter of 200 nm or less exhibited minimal uptake. However, as the particle size increased, the rate of spleen accumulation also increased. The authors concluded that particles between 70 and 200 nm in diameter were optimal for avoidance of liver and spleen uptake. The particle surface charge can further be modulated to minimize opsonization by phagocytic cells of the RES. Levchenko et al. demonstrated that the presence of charged lipids in the liposome bilayer (in the absence of PEG), especially the negatively charged phosphatidic acid (PA) and phosphatidylserine (PS), strongly accelerated the clearance of liposomes from blood [11]. However, the liposome pharmacokinetics becomes more complex if it contains both PEG and a charged phospholipid [11]. More recently, Geng et al. [12] reported that nanoparticle shape can influence their pharmacokinetics and RES sequestration. Specifically, they demonstrated that cylinder-shaped filomicelles were able to achieve blood circulation half-life as long as 5 days, about 10 times greater than their spherical counterparts. Their in vitro macrophage studies revealed that these worm-shaped nanoparticles experience a strong drag force in the presence of fluid flow which minimized macrophage engulfment [12]. However, the feasibility of cylinder-like nanoparticles to carry large loads of imaging and/or therapeutic agents has not yet been shown. As a result, currently available evidence support the

5 development of nanoparticles that are PEGylated, of approximately 100 nm in diameter and have a neutral surface charge for applications requiring prolonged blood circulation lifetime. Tumor targeting via the passive enhanced permeation and retention (EPR) effect, first described by Matsumura and Maeda in 1986 [13], requires particles to have a prolonged vascular residency time (i.e. maintain high plasma AUC for > 6 h in mice and rats [13-15]). It has been agreed that the degree of macromolecule accumulation in tumors is directly proportional to the blood AUC (or exposure) and inversely proportional to the rate of urinary clearance [16-18]. Once the prerequisite of high exposure has been achieved, the transport of macromolecules, such as nanoparticles, into tumor tissues is further affected by the tumor vascular pore size (up to 400 nm, [19]). However, their subsequent intratumoral retention is a function of the particle diffusivity in the tumor interstitial space [20], the speed of the tumor venous return (usually slower than normal tissue [21, 22]), as well as the presence of a poor lymphatic drainage system [21, 22]. Altogether, macromolecules and nanoparticles not only preferentially accumulate in tumors via the enhanced vascular permeability, but they are also preferentially retained there (for multiple hours to days). Conversely, low-molecular-weight agents are distributed systemically following administration, rapidly cleared from the circulating blood via renal clearance, and their tumor accumulation is only transient (on the order of minutes). Their small size allows them to readily return to into the circulating blood system following extravasation into the tumor interstitial space [15, 23]. The ability of the EPR effect to significantly increase tumor accumulation versus healthy tissue distribution results in increased target-to-background signal ratio during imaging and enhanced tumor-tohealthy tissue therapeutic ratio during treatment. EPR is the hallmark of nanoparticle-based delivery of diagnostic and therapeutic agents to tumors [18].

6 Once the tumor site is reached, different strategies have been developed for selectively directing the nanoparticles to target specific cell populations and sub-cellular compartments [24, 25].

These include the engineering of nanoparticles responsive to

different triggers that are either inherently present in the tumor microenvironment (i.e. pH [26], matrix metalloproteinases [27]) or that can be induced externally (i.e. temperature [28, 29], light irradiation [30]). Molecularly targeted surface ligands can also be incorporated onto the outer layer of the nanoparticles. These enable the retention of nanoparticles either on the surface of the cells of interest or induce cellular uptake. Furthermore, appropriate surface modifications also lead to successful targeting to intracellular compartments such as the nucleus or the mitochondria [31-33]. The liposome system developed and employed in this thesis is a passive, sterically stabilized particle that solely relies on the EPR effect to achieve tumor targeting. Extensive characterization of the distribution patterns and kinetics of passive systems is necessary to lay down the groundwork needed for future quantification aimed at assessing advantages of the different active targeting strategies.

1.2.

Rationale for Spatio-Temporal Biodistribution Assessment The pharmacokinetics and biodistribution profile is often used as a surrogate to

evaluate the potential effectiveness of a new therapeutic or diagnostic agent. For example, the blood concentration of a drug has often been correlated to its efficacy and toxicity [1]. As a result, the agent should be designed to reach the desired therapeutic or diagnostic effect at the lowest administration dose possible. The development process therefore aims to select the

7 formulation that yields the highest agent concentration at the desired target site (i.e. tumor) and the lowest agent concentration elsewhere (i.e. healthy organs, background tissues). The temporal component of a biodistribution assessment is also important. In the case of a diagnostic agent, the biodistribution kinetics defines the optimal imaging window for obtaining information on a specific biological or physiological process. For example, in a routine functional CT or dynamic contrast enhanced (DCE)-MR imaging session, it is important to characterize the distribution and clearance kinetics of the agent in order to accurately define the arterial and venous phases. If the imaging probe employed is involved in an active biological process, such as fluorodeoxyglucose (FDG) in cellular metabolism or ultra-small superparamagnetic iron oxide (USPIO) in macrophage phagocytosis, the timelines of these processes must be defined in order to set the time gap between probe administration and imaging (i.e. one hour for FDG-PET, and 24 hours for USPIO-MR). In the case of a therapeutic agent, its pharmacokinetics and biodistribution influence its efficacy and toxicity. For example, studies conducted in mice to compare the efficacy of liposomeencapsulated doxorubicin versus free doxorubicin found a gain in the plasma AUC of at least 60-fold [34-37] and an increase of 14-fold in the peak tumor drug concentration for the liposomal drug [35]. This resulted in an enhancement in treatment efficacy (i.e. 6-fold [38]) and a significant decrease in toxicity [34, 39]. Characterization of the temporal profile of the biodistribution and clearance of the therapeutic agent of interest will better enable the setting of appropriate dosing regimens.

8 1.3.

Imaging as a Non-invasive Method for Nanoparticle Biodistribution Assessment Traditional nanoparticle pharmacokinetics and biodistribution studies rely on plasma

and tissue sampling. The invasive nature of these procedures can change the biological system under observation resulting in unreliable representation of the in vivo conditions. Furthermore, animals often need to be sacrificed at each sampling time point in order to either provide enough plasma or tissue for analysis or because their vital organs are removed during the biodistribution assessment process. These limitations associated with traditional pharmacokinetics and biodistribution studies can be overcome through the use of noninvasive imaging techniques in conjunction with appropriate labeling of the nanoparticle of interest. Image-based measurements, when successfully correlated with tissue agent concentrations, can be used to collect meaningful data on the same animal over multiple time points with minimum amount of perturbation to biological and physiological processes [40]. Therefore, not only animal-to-animal variations are avoided, but also the total number of animals required for each study can be reduced [41]. Furthermore, thanks to the increasing availability of small animal scanners, the imaging assays employed in the preclinical environment can be more readily translated to the clinical setting. To date, non-invasive nuclear imaging techniques such as PET and single photon emission

computed

tomography

(SPECT)

have

been

extensively

explored

for

pharmacokinetics and biodistribution studies [42, 43]. PET isotopes have shown an advantage over SPECT isotopes for radiolabeling of small molecules because atom replacement is possible with positron emitters such as

11

C,

15

O,

13

N and

18

F in a compound

without modification of its pharmaceutical, biological or biochemical properties [44]. However, for long-circulating nanoparticulates such as liposomes, PET labeling is unsuitable

9 because the positron emitters often have much shorter physical half-lives (10 minutes for 13N, 20.4 minutes for 11C, 110 minutes for 18F and 12.7 hours for 64Cu) compared to the vascular half-life of the nanoparticles (~ 55 hours for liposomal doxorubicin, Doxil). As such, PET tracer-labeled pharmaceutical carriers cannot be tracked over their full circulation-lifetime in vivo and repeated imaging over multiple time points is greatly limited. Some SPECT radiotracers have physical half-lives that are suitable for longitudinal carrier therapeutics studies, such as

111

In with t1/2 of 67.9 hours. The presence of a metal chelating agent such as

diethylene triamine pentaacetic acid (DTPA) when labeling a macromolecular agent is also less of a concern compared to its use for the labeling small molecules as it usually does not significantly affect the pharmacokinetics and distribution of the macromolecule [45]. In addition, advanced SPECT systems allow for collection of photons emitted at different energy windows. This is very valuable for applications requiring simultaneous imaging of multiple radiotracers (i.e. monitoring the biodistribution of multiple species of carriers labeled with radionuclides of different and resolvable gamma energies). However, as the radioactivity of the tracer decays, the imaging time need to be increased in order to maintain the image quality and count statistics. Harrington et al. [46] conducted a clinical study administering

111

In-DTPA-labeled liposomes to 17 patients with locally advanced cancers.

Serial whole body gamma camera images were acquired up to 7 days post-injection showing liposome localization in the tumor lesions as well as in healthy tissues. The image quality of the data set acquired at day-7 was significantly deteriorated due to both physical decay of 111

In and biological clearance of liposomes. This report confirms that the performance of

SPECT over the course of a longitudinal biodistribution assessment is not constant. In addition, just like PET, the use of SPECT imaging to map the colloidal carrier tissue

10 distribution is greatly limited by its inability to provide structural and anatomical information. Recent development of integrated imaging systems such as PET/CT and SPECT/CT has enabled acquisition and fusion of anatomy with the nuclear imaging data. Furthermore, the development of PET/MR and SPECT/MR systems will likely provide an even more fertile ground for innovations in the area of nuclear medicine-based monitoring of biodistribution of long circulating nanoparticles. More recently, MR has also been explored as a potential tool to image tissue distribution of nanocarriers [47-49]. Viglianti et al. [50, 51] reported a particularly interesting set of studies where MR was successfully used to visualize and quantify drug release from temperature sensitive liposomes labeled with Mn2+. Advantages of MR over other imaging modalities (i.e. CT, PET and SPECT) include the absence of ionizing radiation and high softtissue contrast. However, its sensitivity for measuring T1 and T2-shortening contrast agent concentrations (i.e. 10-5 M for Mn2+ [51]) is about 105 to 107 times lower than SPECT (10-10 M [52]) and PET (10-11 to 10-12 M [52]) , respectively. In addition, although a number of fast mapping techniques have been described for quantification of T1 [53-55] and T2 [56], data collection times are still generally lengthier and image resolution lower than that of conventional qualitative MR acquisitions [57, 58]. The whole-body imaging techniques described above have the ability to quantify drug tissue distribution non-invasively. Their resolution limitations make them inadequate for providing information on the intracellular localization of the administered agents. Microscopes with fluorescence detectors are currently the most widely used tool for visualization of molecule distribution within a cell [59]. Because optical imaging techniques lack depth penetration and are heavily affected by scattering effects, quantitative in vivo

11 imaging cannot be performed. Hence, if the administered drug and carrier were either inherently fluorescent (i.e. doxorubicin) or labeled with a fluorescent probe, then by collecting a small biopsy sample from the location of interest, the intracellular localization of both the drug and carrier can be visualized using a confocal microscope. However, accurate quantification of fluorescence is sometimes difficult even ex vivo due to optical property changes. For example, doxorubicin fluorescence is partially quenched when the drug binds to the deoxyribonucleic acid (DNA) [60]. The work conducted within this thesis is aimed at characterizing the whole body biodistribution map and kinetics of liposomes using volumetric imaging modalities (CT and MR). This liposome platform has been engineered to allow for future addition of modular components that would support imaging in other modalities such as PET, SPECT and optical. This modular multimodality approach will enable quantification of the whole body distribution and cellular uptake of nanoparticles in vivo across a wide range of spatial resolution and detection sensitivity scales through full exploitation of the respective strengths of different imaging techniques.

1.4.

Thesis Outline A versatile multi-purpose contrast agent system is highly advantageous as it can

provide multi-parametric characterization of the disease following just one single injection. This thesis describes the development and characterization of a novel liposome system with exploration of a number of different applications in cancer. Chapter 2 [61] focuses on the formulation and in vitro characterization of the loading, size, morphology, stability and imaging properties of this liposome agent. It also illustrates the potential use of this system

12 for CT and MR dual-modality imaging during image-guided therapeutic procedures by demonstrating that the signal enhancements in both imaging modalities are co-localized. Chapter 3 [62] describes the pharmacokinetics and biodistribution of the liposome system in healthy mice determined by evaluation using traditional blood sampling and whole organ digestion methods. The pharmacokinetics profile obtained was fitted to a one-compartment model and the volume of distribution of the liposomes was matched to that of the blood volume of an average mouse. The vascular half-life of this liposome system was calculated to be approximately 100-fold greater than that of a clinically available small molecule CT or MR contrast agent. This showed feasibility to employ these liposomes as intravascular contrast agents for longitudinal imaging applications. Chapter 4 [63] explores the suitability of these imageable liposomes to be used in image-guided drug delivery. Volumetric CT methods were used to measure the concentration of liposome carriers in the organs and tissues of VX2-carcinoma bearing rabbits over a 14-day period. It is concluded that CT has the ability to detect in vivo concentrations of iodine at sensitivity as high as 8 nmol/cm3 (equivalent to 1 µg/cm3) while maintaining the ability to identify boundaries of anatomical structures at sub-millimeter resolution. Using this approach, heterogeneity in the intratumoral distribution of the liposomes was visualized and their intratumoral volume of distribution quantified in vivo. As a result, the combined use of iodinated liposomes and CT imaging allows for monitoring of colloidal drug delivery and provides an opportunity for online adjustment of therapeutic regimens and implementation of adaptive pharmaceutical delivery. Chapter 5 investigates the ability of the liposome system developed within the framework of this thesis to reach sites of tumor and inflammation. The performance of liposome-CT is then compared to that of FDG-PET for whole-body detection of suspect lesions in a rabbit model

13 bearing both VX2-carcinoma and immune myositis. Comprehensive histopathology was also conducted to confirm the abnormalities. Liposomes induced contrast enhancement in CT at sites of tumor and inflammation. Interestingly, the mean accumulation of liposomes at the inflammatory lesions, observed at five days post-administration, was significantly higher than at found at the tumor sites (p < 0.0001). The partial volume adjusted maximum standardized uptake values (SUVmax) measured from the FDG-PET data set did not yield significant differences in FDG uptake between the two lesion types (p > 0.15). These observations suggest that this liposome agent could play a potential role in increasing the specificity of disease detection and localization. Finally, Chapter 6 discusses the challenges and opportunities for its ready translation into the clinical setting (i.e. commercialization and application for regulatory approval), potential modifications that would increase its performance and versatility, as well as additional investigations needed to better define its role in image-based characterization of tumor morphology and patho-physiology.

Chapter 2. Multimodal Contrast Agent for Combined CT and MR Imaging Applications

14

15

2.1.

Foreword Innovations in nanoparticle design and construction open opportunities for

engineering of novel imaging agents that carry signal generating moieties for more than one imaging modality – these are referred to as multimodal imaging agents. The following chapter describes the development and characterization of a liposome-based CT and MR contrast agent and it has been published as: Zheng J, Perkins G, Kirilova A, Allen C, Jaffray DA. Multimodal Contrast Agent for Combined Computed Tomography and Magnetic Resonance Imaging Applications. Investigative Radiology, Volume 41, Number 3, Pages 339 – 348. March 2006. It has been reproduced with permission from Lippincott Williams & Wilkins.

2.2.

Introduction In recent years there has been an increase in the use of multimodality imaging (i.e.

CT/PET, CT/SPECT, x-ray/MR) [64-71]. Since each medical imaging modality has unique strengths and limitations, it is often through the compound use of multiple modalities that the complete assessment of a patient is achieved. Interest in the area of multimodality imaging has also been prompted by the realization that such techniques offer much more sophisticated characterization of the morphology and physiology of tissues and organs, and that confidence gained in the accurate correspondence or registration of different modalities greatly enhances their value [72]. This improved value of imaging will ultimately allow for advances in diagnosis and evaluation of disease, image-guided therapeutic interventions, and assessment of treatment outcomes. The recent integration of CT and PET systems is a good example of the advantages of the multimodal approach [64-66]. The CT-PET combination has

16 revolutionized the utilization of PET in diagnostic applications since it has been shown to increase the specificity of PET-based assessment by accurately placing the diseased structure within the body frame [73-75]. In the context of radiation therapy, there is a need to merge CT and MR imaging - CT is employed for 3D volumetric radiation dose calculation and MR is utilized for accurate delineation of the target and normal structures [76]. For example, accurate delineation and targeting of the prostate gland in radiation therapy of prostate cancer necessitates parallel use of CT and MR imaging [77]. Furthermore, CT technology in the form of conventional and cone-beam systems is employed on a daily basis to guide the delivery of radiation therapy on treatment machines [78, 79]. The development of a multimodal CT and MR contrast agent with the ability to facilitate target delineation and assist in the guidance of therapy has the potential to increase both the accuracy and the precision of the delivery of radiation therapy. Clinical imaging in all modalities requires that an adequate level of differential contrast relative to noise be achieved in order to identify the structures or phenomena under observation. Although imaging on CT and MR can be performed without the administration of contrast agents there are numerous instances in both disease diagnosis and treatment, in which procedures benefit from the improved contrast and dynamics that are added by the use of these agents [80, 81]. In addition, if the multimodal agent’s localization in the body is persistent enough, it can potentially become a relatively non-invasive alternative to fiducial markers for image-guided radiotherapy procedures. Given these considerations, it is the objective of these investigations to develop an agent to assist in the multimodal registration process through the creation of spatially consistent image signals across CT, MR and conebeam CT for radiation therapy applications.

17 The present study proposes unilamellar liposomes as a delivery system (Figure 2.1) for two commercially available contrast agents: Omnipaque (iohexol, Nycomed Imaging AS, Oslo, Norway), a CT agent, and ProHance (gadoteridol, Bracco Diagnostics Inc., Princeton, NJ, USA), an MR agent. The objective of this study is to examine the feasibility of such a multimodal system to effectively induce and maintain contrast enhancement in both CT and MR. Specifically, the size, morphology and encapsulation efficiency of the liposomes for both CT and MR agents are measured. The in vitro stability of the system and in vitro release kinetic profiles of the encapsulated agents are determined. The relaxivity characteristics and the in vitro CT and MR imaging properties of the system are investigated in a phantom. In addition, a preliminary imaging-based assessment of the in vivo stability of this multimodal contrast agent is conducted in a lupine model. These series of studies represent the first step towards the development of a colloidal carrier-based multimodal contrast agent for combined CT and MR imaging.

18

Figure 2.1 Schematic of the liposome-based contrast agent system (not drawn to scale).

19

2.3.

Materials and Methods

Materials The

components

of

liposomes:

1,2-Dipalmitoyl-sn-Glycero-3-Phosphocholine

(DPPC, M.W. 734), Cholesterol (CH, M.W. 387) and 1,2-Distearoyl-sn-Glycero-3Phosphoethanolamine-N-[Poly(ethylene glycol)2000] (PEG2000DSPE, M.W. 2774) were purchased from Northern Lipids Inc. (Vancouver, British Columbia, Canada).

The CT

contrast agent, Omnipaque was obtained from Nycomed Imaging AS, Oslo, Norway. Omnipaque (300 mg/mL of Iodine) contains iohexol (M.W. 821.14), an iodinated, watersoluble, non-ionic monomeric contrast medium. The MR contrast agent used was ProHance from Bracco Diagnostics Inc. (Princeton, NJ, USA). ProHance (78.6 mg/mL of gadolinium) contains gadoteridol (M.W. 558.7), a non-ionic gadolinium complex of 10-(2-hydroxypropyl)-1,4,7,10-tetraazacyclododecane-1,4,7-triacetic acid.

Preparation of liposome formulations Lipid mixtures (200 mmol/L) of DPPC, cholesterol and PEG2000DSPE in 55:40:5 percent mole ratios were dissolved in ethanol at 70°C. The lipid-ethanol solution was then hydrated at 70°C with Omnipaque (300 mg/mL of iodine, 45%vol) and Prohance (279.3 mg/mL of gadoteridol, 45%vol). The initial ethanol content was 10%vol. The resulting multilamellar vesicles were then extruded [82, 83] at 70°C with a 10 mL LipexTM Extruder (Northern Lipids Inc., Vancouver, British Columbia, Canada). Specifically, the samples were first extruded 5 times with two stacked polycarbonate membranes of 0.2 µm pore size

20 (Nucleopore Track-Etch Membrane, Whatman Inc., Clifton, NJ, USA) and subsequently 5 times with two stacked polycarbonate membranes of 0.08 µm pore size.

Physico-chemical characterization of liposome formulations Liposome size and morphology The size of liposomes was measured by dynamic light scattering (DLS) at 25°C using a DynaPro DLS (Protein Solutions, Charlottesville, VA, USA). Liposome morphology was studied by transmission electron microscopy (TEM) with a Hitachi 7000 microscope operating at an acceleration voltage of 80 kV. The liposome sample was first diluted in distilled water and then mixed with phosphotungstic acid (PTA) in a 1:1 volume ratio. The sample solutions were then deposited onto negatively charged copper grids that had been pre-coated with carbon.

Evaluation of loading efficiency, in vitro stability and in vitro release kinetics Following liposome preparation, the unencapsulated agent was removed by membrane dialysis. Specifically, 1 mL of the liposome sample was placed in an 8000 molecular weight cut-off (MWCO) dialysis bag suspended in 250 mL of N-(2Hydroxyethyl)Piperazine-N'(Ethanesulfonic Acid) (HEPES) buffer saline (HBS) and left to stir for 8 hours. The liposomes were then ruptured using a 10-fold volume excess of ethanol in order to measure the concentration of encapsulated agents. The iodine concentration was determined using a UV assay with detection at a wavelength of 245 nm (Heλios γ, Spectronic Unicam, MA, USA). The gadolinium concentration was determined using an assay based on inductively coupled plasma atomic emission

21 spectrometry (ICP-AES Optima 3000DV, Perkin Elmer, MA, USA). The encapsulation efficiency of the agents was calculated using the following equation: % encapsulation efficiency =

amount of agent encapsulated ⋅ 100 amount of agent added during preparation

The in vitro release kinetic profile for both agents was assessed by the dialysis method [84]. In short, 1 mL of the liposome sample was placed in a dialysis bag (MWCO 8000) suspended in 250 mL of HBS and incubated at 4°C or 37°C. At specific time points, 5 mL of the dialysate was removed for measurement of the iodine and gadolinium concentrations and 5 mL of fresh HBS was added in order to maintain constant volume. The stability of the liposomes was assessed by measuring the size of liposomes at specific time points during the incubation period.

In vitro CT and MR imaging CT scanning was performed using a GE LightSpeed Plus 4-detector helical scanner (General Electric Medical Systems, Milwaukee, WI, USA) with the following scan parameters: 2.5 mm slice thickness, 120 kV, 300 mA and 15.2 x 15.2 cm field of view (FOV). The mean attenuation in Hounsfield units (HU) was measured in each agentcontaining tube (1 cm in diameter) using circular regions of interest (ROI) of 7 mm2. MR imaging was performed with a head coil in a 1.5 Tesla GE Signa TwinSpeed MR scanner (General Electric Medical Systems, Milwaukee, WI, USA). Scans were produced using a T1 weighted spin echo sequence with a repetition time (TR) of 450 ms, an echo time (TE) of 9 ms, a slice thickness of 3 mm, a FOV of 19.9 x 19.9 cm and an image matrix of 256 x 192 pixels. The relative signal intensity was taken over an ROI of 7 mm2.

22 In vitro relaxometry All in vitro relaxometry measurements were performed at 20°C on a 1.5 Tesla, 20cm-bore superconducting magnet (Nalorac Cryogenics Corp., Martinez, CA) controlled by an SMIS spectroscopy console (SMIS, Surrey, UK). The T1 relaxation time data were acquired using an inversion recovery (IR) sequence [85] with 35 inversion recovery time (TI) values logarithmically spaced from 1 to 32000 ms. A 10 second delay was given between each acquisition and the next inversion pulse. The T2 relaxation time data were acquired using a CPMG sequence [85, 86] with TE/TR = 1/10000 ms. For every measurement 2000 even echoes were sampled with 8 averages. The effects of any residual transverse magnetization following the off-resonance irradiation was removed by phase-cycling the π/2 pulse (x/x). The T1 relaxation data were analyzed assuming mono-exponential behavior t −   ( S = M 0 ⋅ 1 − 2 ⋅ e T 1  , where S is the signal observed, M0 is the magnetization at  

equilibrium, t is time and T1 is the longitudinal relaxation time). All T2 decay data were plotted to a one component T2 model with a Gaussian fit on a logarithmic time scale. The r1 and r2 values were calculated from linear regression analysis of 1/T1 and 1/T2 relaxation rates versus gadolinium concentration.

In vivo CT and MR imaging The in vivo imaging study was performed under a protocol approved by the Animal Care and Use Committee of the University Health Network. A New Zealand white rabbit (male, 3.5 kg) was anaesthetized with an intramuscular injection of 40 mg/kg of ketamine

23 and 5 mg/mL of xylazine. 2% isoflurane vapor was given by inhalation throughout the study. 10 mL of the liposome-based contrast agent solution (36 mg/kg of iodine and 12 mg/kg of gadolinium) was injected into the marginal ear vein catheter at a rate of 1 mL/second. Preand post-contrast injection images of the rabbit were acquired in both imaging modalities. 15, 60, 120, 180 minutes and 24, 72 and 168 hours following the contrast agent injection, the rabbit was imaged in CT (120kV, 200mA, FOV = 22.0 x 22.0 cm, slice thickness = 1.25 mm and image matrix of 512 x 512) and then moved to the MR scanner at 30, 90, 150, 200 minutes and 24, 72 and 168 hours post-contrast to acquire images in MR (3D FSPGR sequence with a TR of 8.5 ms, a TE of 4.1 ms, a slice thickness of 3.0 mm with an overlap of 1.5 mm, an FOV of 22.0 x 22.0 cm and an image matrix of 256 x 256). The signal intensity in MR and the mean attenuation values (HU) in CT were measured over a circular ROI of 4 mm2. The cross-sectional images were exported from a review station (Merge eFilm, Milwaukee, WI, USA). The same window and level was used for the pre and post-contrast images.

2.4.

Results

Physico-chemical characterization of liposome formulation The prepared liposome formulation resulted in vesicles having a spherical morphology (Figure 2.2) and a mean diameter of 74.4 ± 3.3 nm. Table 2.1 summarizes the agent loading properties of the liposome formulation. The average loading efficiency (n=8) achieved for iohexol was 19.6 ± 2.8 % (26.5 ± 3.8 mg/mL iodine loaded, approximately 2.4x104 iohexol molecules per liposome), which represents an agent to lipid ratio of approximately 0.2:1 (wt:wt). The average loading efficiency (n=8) attained for gadoteridol

24 was 18.6 ± 4.4 % (6.6 ± 1.5 mg/mL gadolinium loaded, approximately 1.4x104 gadoteridol molecules encapsulated in one liposome), which represents an agent to lipid ratio of approximately 0.05:1 (wt:wt).

Figure 2.2 Transmission electron micrograph of the negatively stained dual-agent containing liposomes at (a) 40,000 magnification and (b) 80,000 magnification.

25

Diameter (nm)

Iodine added (mg/mL)

Iodine loaded (mg/mL)

Iodine loading efficiency (%)

Gadolinium added (mg/mL)

Gadolinium loaded (mg/mL)

Gadolinium loading efficiency (%)

74.4 ± 3.3

135

26.5 ± 3.8

19.6 ± 2.8

35.5

6.6 ± 1.5

18.6 ± 4.4

Table 2.1 Size and loading characteristics of the dual-agent-containing liposome formulation (n = 8). The liposomes are composed of DPPC/cholesterol/DSPE-PEG (55/40/5/ mole ratio). Data represent the mean ± standard deviation.

Figure 2.3 includes the in vitro release profile for both agents under sink conditions in physiological buffer at 4ºC (Figure 2.3a) and 37ºC (Figure 2.3b). As shown, following the 15-day incubation period at 4°C, 8.7 ± 1.5 % and 6.6 ± 4.5 % of the encapsulated iodine and gadolinium were released, respectively, and at 37°C, 9.1 ± 2.5 % and 7.5 ± 1.4 % of the encapsulated iodine and gadolinium were released, respectively. The liposomes were also sized periodically during the incubation period in order to assess their stability under sink conditions in HBS at 37ºC. As seen in Figure 2.4 the liposome size remains constant throughout the incubation period.

In vitro imaging Visual contrast enhancement was observed in CT and MR when the liposome-based contrast agent was imaged in vitro at varying concentrations (Figures 2.5a and 2.5b). Figure 2.6a illustrates the measured CT attenuation of the liposome encapsulated contrast agents, the unencapsulated iohexol, the unencapsulated gadoteridol and the mixture of unencapsulated

26 (a)

(b)

Figure 2.3 The in vitro release profile for iohexol and gadoteridol from DPPC/cholesterol/DSPE-PEG (55/40/5 mole ratio) liposomes dialyzed under sink conditions (250-fold volume excess) against HBS (a) at 4 °C and (b) at 37 °C (n = 4). Data are represented as the mean ± standard deviation.

27 iohexol and gadoteridol. Attenuation values varied linearly with concentration for all contrast agent solutions. Linear regression analysis revealed an attenuation of 38.5 ± 0.5 HU/(mg of gadolinium) in 1 mL of HBS for the unencapsulated gadoteridol (r=0.99), 29.0 ± 0.4 HU/(mg of iodine) in 1mL of HBS for the unencapsulated iohexol (r=0.99), 37.8 ± 0.5 HU/(mg of iodine and 0.2 mg of gadolinium) in 1 mL of HBS for the mixture of unencapsulated iohexol and gadoteridol (r=0.99), and 36.3 ± 0.5 HU/(mg of iodine and 0.2 mg of gadolinium) in 1 mL of HBS for the liposome formulation (r=0.99). The slightly lower attenuation values observed for the liposome encapsulated iohexol and gadoteridol compared to free iohexol and gadoteridol are due to the presence of lipids, which, with respect to water, have lower CT attenuation values.

Figure 2.4 Size of the dual-agent-containing liposomes during dialysis under sink conditions (250-fold volume excess) against HBS at 37 °C (n = 3). Data are represented as the mean ± standard deviation.

28 (a)

(b)

Figure 2.5 In vitro imaging efficacy of the liposome-based contrast agent system (a) in CT (2.5 mm slice thickness, 120 kV, 300 mA and 15.2 cm2 FOV) and (b) in MR (450 ms TR, 9 ms TE, 3 mm slice thickness, 19.9 cm2 FOV and 256 x 192 image matrix). Data are represented as the mean ± standard deviation.

Figure 2.6b illustrates the MR relative signal profile as a function of gadolinium or iodine concentration. It is known that the relationship between gadolinium concentration and relative signal intensity in MR becomes markedly non-linear at high concentrations of gadolinium [87-89]. Furthermore, negative enhancement occurs in MR when the gadolinium concentration reaches high enough levels to cause significant T2 shortening, which in turn results in signal loss [90-93]. The plots in Figure 2.6b for liposome encapsulated gadoteridol and iohexol, free gadoteridol and iohexol, liposome-encapsulated gadoteridol and free

29 gadoteridol all exhibit non-linear characteristics. The free iohexol plot confirms that iodine in the concentration range of 0 to 17 mmol/L shows signal intensity levels comparable to those achieved by water. The average differential signal intensity (SI) in MR for free iohexol samples was 1.8 ± 7.1 SI relative to water. The unencapsulated gadoteridol samples reached peak differential signal intensities (> 600 SI with respect to water) in the gadolinium concentration range of 1 to 9 mmol/L. This is in accordance with previous findings [89, 92]. A slight decrease in the mean signal intensity was observed when free gadoteridol was mixed with iohexol. This finding is consistent with previous reports on the capability of iodinated contrast agents to diminish the signal enhancing effects of gadolinium [94-96]. Encapsulation of gadoteridol in liposomes (in the presence and absence of iohexol) was found to cause a right shift in the differential signal intensity profile (peak signal intensities in MR achieved with gadolinium concentration ranging from 5 to 18 mmol/L). Encapsulation of gadoteridol in the interior of liposomes diminishes MR signal at lower gadolinium concentrations (< 5 mmol/L) due to limited bulk water access which decreases 1/T1 values [97]. At higher gadolinium concentrations (> 5 mmol/L), however, encapsulation of gadoteridol significantly dampens the T2 relaxation effect allowing high signal levels to be maintained over a much broader gadolinium concentration range in MR.

30

(a)

(b)

Figure 2.6 (a) CT (2.5 mm slice thickness, 120 kV, 300 mA and 15.2 cm2 FOV) attenuation in HU as a function of contrast agent concentration in mmol/L. Although gadolinium has CT attenuation properties, iodine provides more effective CT enhancement. (b) Differential signal intensity (with respect to water) in MR (450 ms TR, 9 ms TE, 3 mm slice thickness, 19.9 cm2 FOV and 256 x 192 image matrix) as a function of increasing gadolinium and iodine concentrations. Symbols represent liposome-encapsulated gadoteridol and iohexol (■), liposome-encapsulated gadoteridol (●), free iohexol and gadoteridol (▲), free gadoteridol (●) and free iohexol (▼). Data are represented as the mean ± standard deviation.

31 In vitro relaxometry For the relaxometry measurements, T1 (Figure 2.7a) and T2 (Figure 2.7b) rates were observed to be linear and concentration dependent for both the liposome encapsulated and the unencapsulated contrast agents. The r1 and r2 values of unencapsulated gadoteridol were 5.1 and 6.2 s-1mmol-1L, respectively. The r1 and r2 values for gadoteridol in the presence of iohexol were 6.4 and 7.8 s-1mmol-1L, respectively, and the r1 and r2 values for the liposome encapsulated agents were 1.2 and 1.5 s-1mmol-1L. The r1 and r2 values for iohexol were found to be 0.0 s-1mmol-1L. Therefore, the encapsulation of the paramagnetic agent gadoteridol in liposomes (in the presence of iohexol) significantly reduces both the 1/T1 and 1/T2 relaxivity values, in accordance with Figure 2.6b, as well as previously published data [97].

In vivo imaging Preliminary in vivo imaging shows visual contrast enhancement in the heart (Figure 2.8) and major blood vessels in both CT and MR up to 72 hours (3 days) following administration of the liposome-based multimodal contrast agent. Figure 2.9 illustrates the maintained measurable signal enhancement found in the blood (measured in the aorta) for the two imaging modalities. Specifically, the signal intensities in MR were increased by over 200% after the administration of the multimodal contrast agent for 72 hours and then decreased to signal intensities that were approximately twice as high as the pre-contrast injection values 7 days post administration. In CT, a 60% increase in HU was achieved and maintained for 3 hours following administration of the agent and a 35% increase in HU was detectable at 72 hours post-contrast injection. No measurable increase in HU was found 7 days post-contrast injection. The prolonged enhancement achieved in the blood pool in both

32 imaging modalities demonstrates that the liposome carriers are able to circulate and reside in the blood while retaining the co-encapsulated small molecular weight agents.

(a)

(b)

Figure 2.7 (a) 1/T1 relaxation rate and (b) 1/T2 relaxation rate as a function of gadolinium (Gd) and iodine (I) concentration obtained at 20°C with a 1.5T, 20-cm-bore superconducting magnet controlled by an SMIS spectroscopy console. Encapsulation of gadoteridol greatly reduces both the r1 and r2 of the gadolinium atoms. Symbols represent free gadoteridol (■), free iohexol and gadoteridol (●), free iohexol (▲) and liposome encapsulated agents (▼). The r1 and r2 values for all four solutions are listed in Table 2.2. Data are represented as the mean ± standard deviation.

33

r1 (s-1mmol-1L)

r2 (s-1mmol-1L)

Free gadoteridol

5.14 ± 0.06

6.21 ± 0.08

Free gadoteridol and iohexol (1:29 mole ratio of Gd to I)

6.38 ± 0.16

7.83 ± 0.20

Free iohexol (x-axis = [I] in mmol/L)

0.00 ± 0.00

0.01 ± 0.01

Liposome encapsulated agents

1.23 ± 0.02

1.46 ± 0.02

Table 2.2 Relaxivity r1 and r2 values for the free gadoteridol, free iohexol and gadoteridol, free iohexol and liposome encapsulated agents solutions plotted in Figures 2.7a and 2.7b. Data are represented as the mean ± standard deviation.

Figure 2.8 Illustration (not quantitative) of the use of the liposome-based contrast agent (single administration of 36 mg/kg of iodine and 12 mg/kg of gadolinium co-encapsulated in liposomes) in a 3.5 kg white New Zealand rabbit in CT and MR. CT (120 kV, 200mA) and MR (3D FSPGR sequence, TR/TE=8.5/4.1) axial images at the level of the rabbit heart were obtained before and after contrast agent injection (15, 60, 120, 180 minutes and 24, 72, 168 hours post-contrast in CT and 30, 90, 150, 200 minutes and 24, 72, 168 hours post-contrast in MR). The same window and level were used for pre- and post-contrast injection images. Note the visual contrast enhancement obtained and maintained in the heart in both imaging modalities.

34

35

Figure 2.9 Relative percentage signal enhancement achieved in the aorta of the rabbit measured from MR and CT images using circular regions of interest. In MR, a relative signal intensity increase of 1930.3 ± 188.1 was measured 30 minutes post-contrast injection and a relative signal intensity increase of 1028.5 ± 169.3 was measured 7 days post-contrast injection. In CT, a relative HU increase of 39.2 ± 8.9 was measured 15 minutes post-contrast injection and no measurable HU increase was found 7 days post-injection. Data are represented as the mean ± standard deviation.

36

2.5.

Discussion Rational design of a multimodal contrast agent is a complex endeavour in that

different underlying physical mechanisms are responsible for contrast generation across imaging modalities. In the case of CT, agents containing elements with high atomic number, such as iodine, are able to increase the differential x-ray attenuation between different soft tissues and organs. Whereas, MR contrast agents made up of paramagnetic metals, such as gadolinium, are able to deliver signals by increasing surrounding tissue relaxivity. Furthermore, the differences in intrinsic sensitivity and resolution between the two imaging modalities create a requirement for substantially different concentrations of each reporter moiety in order to achieve adequate signal intensity1. For example, in a clinical context, MR is sensitive to gadolinium concentrations between 1-10 µg/mL, while CT requires at least 1 mg/mL of iodine for detection [80]. A multimodal contrast agent with efficacy in CT and MR must, therefore, accommodate this 100-fold differential in sensitivity and minimize any agent-related signal interferences across different imaging modalities. To date, although a multitude of contrast agents are commercially available for single modality imaging, few attempts have been made to develop contrast agents that can be used across multiple imaging modalities [100-105]. The lack of development in this area is likely due to challenges presented by the fact that the distinct imaging modalities have distinct sensitivities for different contrast agents [80]. A simple approach for realizing a multimodal contrast agent for CT and MR has been to exploit commercially available extracellular 1

It is important to note that the different physical processes involved in the generation of CT and MR signals contribute to the difference in detection sensitivity for their respective contrast agents (i.e. iodine and gadolinium). For example, for a CT scanner operating at 120 kVp and 200 mA, the photon fluence measured at 50 cm away from the x-ray source is in the order of 108 to 109 photons/mm2 [98]. This means that an iodine atom (of ~ 10-8 mm2 in surface area) has a probability of interacting with only 1 to 10 x-ray photons over the entire exposure time. Conversely in MR, a gadolinium chelate can interact with approximately 106 water protons in one second [99].

37 gadolinium-based contrast agents for enhancement in both of these modalities. In this case, the properties of gadolinium that allow for use in both CT and MR include its relatively high atomic number and paramagnetic characteristics [100-104]. However, due to their low molecular weight, these agents only remain in the vascular system for a short period of time, exhibit rapid dynamic distribution changes in different organs and are excreted quickly. The use of these agents for cross-modality imaging would therefore require both multiple administrations and fast imaging sequences. Also, the low gadolinium payload per molecule, relative to conventional iodinated contrast agents, would necessitate the administration of higher doses for adequate CT enhancement which may have implications in terms of both cost and toxicity [100-104]. Furthermore, the short in vivo residence time of these agents would impose limitations on the size of the anatomic region that could be imaged optimally and would exclude them from being used in image-guidance applications due to their inability to provide prolonged contrast enhancement over the entire course of treatment [81]. An approach to effectively deliver the required amount of contrast in each imaging modality and to prolong the presence of the agents in vivo is to employ particulate carriers such as liposomes. Specifically, liposome-based systems have been evaluated for either encapsulating [106-122] or chelating [123-127] single CT or MR contrast agents. In this study liposomes were selected as the system of choice for delivery of CT and MR contrast agents at appropriate concentrations. The strategy of co-encapsulating two agents in a liposome was pursued for the following reasons: (i) guarantee of consistent transport and distribution of both agents; (ii) liposomes have well understood and characterized physical and biological properties, and formulations based on this technology, such as Doxil (Ortho Biotech Products, L.P., Bridgewater, NJ, USA), DaunoXome (Gilead

38 Sciences, Inc., Foster City, CA, USA) and Nyotran (Aronex Pharmaceuticals, Inc., The Woodlands, TX, USA), have received regulatory approval for clinical use; (iii) both the CT and MR contrast agents selected for encapsulation have been widely used in clinical applications; (iv) encapsulation of iohexol in liposomes does not affect the CT attenuation capability of this agent; therefore, as long as a sufficient quantity of iodine is loaded into the interior of the liposomes adequate signal enhancement is expected; and (v) although gadolinium relaxation is greatly dependent on the amount of water that the gadolinium atoms can access when encapsulated, the permeability of the liposome membrane can be easily adjusted by varying the lipid composition and cholesterol content [128-130]. In the present study, the liposomes were prepared from DPPC, cholesterol and PEG2000DSPE (55:40:5 percent mole ratios). A high cholesterol content (> 40%) was used in order to produce fluid membranes with a high degree of mechanical stability [128, 131]. The fluidity of the membrane will allow for adequate interaction between the encapsulated gadolinium atoms and the external aqueous environment. Cholesterol-rich liposomes (> 40%) have also been shown to be less subject to protein binding when compared to cholesterol-poor (< 20%) liposomes [132, 133]. Furthermore, cholesterol-rich liposomes formed primarily from DPPC are known to be more resistant to the destabilizing effects of serum proteins and have reduced uptake by the monophagocytic system (MPS), when compared to DPPC-based cholesterol-poor formulations [133-135]. The addition of PEG onto the liposome surface is aimed to increase its in vivo circulation lifetime [136, 137]. The presence of PEG will also improve the MR imaging performance since it has been found that liposomes containing 5%mol of PEG can achieve up to two times higher r1 relaxivity values in solution relative to conventional (non-PEGylated) liposomes. This increase in the r1 relaxivity values for the

39 PEGylated liposome solution has been attributed to the presence of PEG-associated water protons in the vicinity of the liposome membrane [138]. The present formulation was prepared using the high-pressure extrusion method [82, 83] and was comprised of spherical vesicles (Figure 2.2) of ~ 74 nm in diameter. This vesicle size was chosen because small unilamellar liposomes of less than 100 nm in diameter have been found to have prolonged in vivo circulation lifetimes [130]. The ideal system for delivery of long circulating contrast agents will have minimal agent release in vivo. A stable formulation with slow release profiles for both agents will allow for prolonged imaging studies and repeated scans in CT and MR. It is known that extracellular agents with small molecular weights such as iohexol and gadoteridol have a much faster clearance profile in vivo compared to colloidal carriers such as liposomes [81]. Therefore, as the encapsulated agents are released from the liposomes, the signal enhancement will diminish in both CT and MR at a rate that is proportional to that of agent release and clearance. In this way, the slow agent release profiles (< 9% of each agent released over 15 days, Figure 2.3) and stability (liposome size remained unchanged over 15 days, Figure 2.4) achieved in vitro for the current liposome formulation has translated into prolonged signal enhancement in vivo in both imaging modalities (Figures 2.8 and 2.9). These studies only include a preliminary evaluation of this system thus a more detailed analysis of the in vivo performance of this agent is a topic of ongoing investigation [139]. The loading characteristics of the current system (Table 2.1) represent roughly 10% of the iodine and gadolinium concentrations found in the commercially available preparations (i.e. Omnipaque and Prohance). However, agents encapsulated in a colloidal delivery system of 74 nm in diameter will have ~1/3 of the volume of distribution in vivo

40 compared to small extracellular agents since the latter will readily cross the fenestrations in the epithelial lining of the blood vessels and enter the interstitium following an intravenous injection [140, 141]. Consequently, directions for future investigations are aimed at achieving higher agent loading levels (~3x) in order to minimize the injection volume required. In addition, there is interest in modifying the surface of the liposomes in order to actively target the vehicles for delivery of agents to specific sites for functional and molecular imaging applications. This proof of principle study has demonstrated the feasibility of engineering a stable liposome-based system that can be used for CT and MR imaging. This system leverages existing clinical knowledge and experience since it employs contrast agents and colloidal carrier technology that are currently approved for use in humans. However, it is necessary to analyze the impact of the liposome formulation on the pharmacokinetics and biodistribution of the agents. These studies, as well as further analysis of the imaging efficacy of the current formulation are underway. Successful optimization of this system may allow for an accelerated development and approval timeline for clinical evaluation. In general, research into the development of multimodal contrast agents has the potential to lead to solutions for the combined challenge of disease detection, treatment design and therapy guidance.

2.6.

Acknowledgements The authors would like to acknowledge Dr. Tom Purdie for operation of the CT

scanner, Dr. Greg Stanisz for use of the SMIS spectroscopy console and T2 data analysis software, Ewa Odrobina for assistance in relaxivity data collection, Jubo Liu for TEM image acquisition and Sandra Lafrance for animal care.

Chapter 3. In Vivo Performance of a Liposomal Vascular Contrast Agent for CT and MR-Based Image Guidance Applications

41

42

3.1.

Foreword The previous chapter described the feasibility of employing a nanoparticle liposome

system to stably co-encapsulate two distinct imaging agents and provide signal enhancement in both imaging modalities. This chapter investigates the in vivo performance of this multimodality liposome system, including its ability to remain confined within blood vessels in healthy animals, its increased vascular circulation life-time compared to administrations of un-encapsulated agents, and the range of intravascular iodine and gadolinium concentration that are quantifiable non-invasively using CT and MR, respectively, and validated by chemical analysis methods through plasma sampling. The overall goal is to demonstrate the utility of this liposome agent for vascular imaging and longitudinal image-guidance applications. The following chapter has been published as: Zheng J, Liu, J, Dunne M, Jaffray DA, Allen C. In Vivo Performance of a Liposomal Vascular Contrast Agent for CT and MR-Based Image Guidance Applications. Pharmaceutical Research, Volume 24, Number 6, Pages 1193 – 1201. June 2007. It has been reproduced with kind permission from Springer Science + Business Media.

3.2.

Introduction There has been a tremendous growth in the use of non-invasive imaging techniques

for characterization of biological processes, diagnosis of disease and guidance of interventions or treatment. Examples of image guided interventions include x-ray, MR and ultrasound-guided surgical procedures [142-145], as well as cone-beam CT-based guidance of radiation therapy delivery [78, 79]. Although different imaging techniques are able to

43 detect inherent contrast in biological systems, conventional diagnostic agents have been employed to enhance soft tissue contrast [146, 147]. However, these are typically low molecular weight molecules and their rapid clearance creates the need for multiple administrations (i.e. angiography). The development of a long circulating contrast agent would offer benefits for guiding interventions in which multiple injections are not feasible or the imaging procedure requires more persistent signal enhancement. Specifically, in radiation therapy, volumetric CT and MR data sets are first acquired and registered for the purpose of radiation dose calculation and target definition [76], and cone-beam CT is then used to guide the delivery of radiation at each treatment session [78, 79]. In this application, the contrast agent is required to provide prolonged signal enhancement for planning (CT and MR), as well as visibility during the process of cone-beam CT acquisition. Thus, an agent with an in vivo lifetime of several days or even weeks would be ideal. A viable strategy to achieve prolonged signal enhancement in vivo is to employ colloidal vehicles to carry conventional contrast agents. Indeed, nano-sized contrast agents have been engineered using liposomes [61, 106, 108, 148-154], lipid and polymeric micelles [155-159], nanoparticles [160-165], dendrimers [166-169] and proteins [170, 171] as carrier systems. In a few cases, these systems have been designed to provide simultaneous contrast enhancement in multiple modalities [61, 105, 150, 172, 173]. However, none of the colloidal systems reported to date have been demonstrated to provide satisfactory and simultaneous signal enhancement in CT and MR. Also, the limited in vivo stability and circulation lifetime of these systems prevent their use throughout both the planning and delivery of radiation therapy.

44 In a previous report, our group summarized the development and in vitro characterization of a dual modality contrast agent for imaging in CT and MR [61]. The agent consists of liposomes co-encapsulating iohexol, an iodine-based conventional CT agent, and gadoteridol, a gadolinium-based conventional MR agent within their internal aqueous compartment. The liposome-based system exhibited high stability in vitro, with less than 10% of the total amount of the encapsulated agents (i.e. iohexol and gadoteridol) released over a 14-day period in physiological buffer at 37°C. The present study is aimed at investigating the in vivo pharmacokinetics and imaging characteristics of this dual modality agent. Specifically, the in vivo stability was evaluated by measuring the pharmacokinetics and biodistribution of the liposome encapsulated CT and MR agents in Balb-C mice following intravenous (i.v.) administration. Studies evaluating the in vivo imaging efficacy were conducted in New Zealand White rabbits using clinical CT and MR scanners. In addition, the signal increases measured in a region of interest in the rabbit aorta in the two imaging modalities were correlated with the actual iodine and gadolinium concentrations detected in plasma samples in order to investigate the potential of using this agent for quantitative imaging applications.

3.3.

Materials and Methods

Materials 1,2-Dipalmitoyl-sn-Glycero-3-Phosphocholine

(DPPC,

Distearoyl-sn-Glycero-3-Phosphoethanolamine-N-[Poly(ethylene

M.W.

734),

glycol)2000]

and

1,2-

(DSPE

-

PEG2000, M.W. 2774) were purchased from Genzyme Pharmaceuticals (Cambridge, MA, USA). Cholesterol (CH, M.W. 387) was purchased from Northern Lipids Inc. (Vancouver,

45 British Columbia, Canada). The CT contrast agent Omnipaque (Nycomed Imaging AS, Oslo, Norway) has an iodine concentration of 300 mg/mL and consists of the non-ionic, iodinated molecule iohexol (N , N ´-Bis(2,3-dihydroxypropyl)-5-[N-(2,3-dihydroxypropyl)acetamido]-2,4,6-triiodo-isophthalamide, M.W. 821.14, 3 iodine atoms per molecule) dissolved in an aqueous solution with tromethamine and edentate calcium disodium. The MR contrast agent ProHance (Bracco Diagnostics Inc., Princeton, NJ, USA) has a gadolinium concentration of 78.6 mg/mL and consists of the non-ionic, gadolinium complex gadoteridol (10-(2-hydroxy-propyl)-1,4,7,10-tetraazacyclododecane-1,4,7-triacetic acid, M.W. 558.7, 1 gadolinium atom per complex) dissolved in an aqueous solution with calteridol calcium and tromethamine.

Preparation and characterization of liposome formulations Liposomes composed of DPPC, cholesterol and PEG2000DSPE in 55:40:5 percent mole ratios were prepared according to a method described in detail elsewhere [61]. Briefly, 100 mmol/L of the lipid mixture was first dissolved in an initial ethanol volume corresponding to 10% of the desired final sample volume. Omnipaque and Prohance were then added to the lipid mixture at a volume ratio of 4:1 and left to hydrate at 75°C for at least 4 hours. The resulting multilamellar vesicles were then sized to 70-85 nm in diameter using high pressure extrusion (10 extrusion cycles) at 70°C with a 10 mL LipexTM Extruder (Northern Lipids Inc., Vancouver, British Columbia, Canada). The un-encapsulated iohexol and gadoteridol molecules were removed by membrane dialysis (8,000 molecular weight cutoff) for 8 hours against 250-fold excess volume of N-(2-hydroxyethyl)piperazineN'(ethanesulfonic acid) (HEPES) buffer saline (HBS). The size of the liposomes was

46 measured by dynamic light scattering (DLS) analysis of dilute solutions using a DynaPro DLS instrument (Protein Solutions, Charlottesville, VA, USA) at 25°C. The final concentration of iohexol was determined using a UV assay with detection at a wavelength of 245 nm (Heλios γ, Spectronic Unicam, MA, USA). The final concentration of gadoteridol was determined using an assay based on inductively coupled plasma atomic emission spectrometry (ICP-AES Optima 3000DV, Perkin Elmer, MA, USA) [61].

Pharmacokinetics and biodistribution studies The pharmacokinetics and biodistribution studies were performed under protocols approved by the University Health Network Animal Care and Use Committee. Female BalbC mice (8-12 weeks, 18-23 g) were administered slow bolus tail vein injections of 150 µL of the contrast agent. Each mouse received 650 mg/kg of iohexol (equivalent to 300 mg/kg iodine) and 60 mg/kg gadoteridol (equivalent to 17 mg/kg of gadolinium) either as a mixture of free agents diluted in HBS or co-encapsulated in liposomes. The animals were anaesthetized with 2% isoflurane and a terminal blood volume (0.5-1.0 mL) was drawn by cardiac puncture at 5, 15, 30 minutes and 1, 2 and 3 hours following the administration of the free agent mixture, and at 5 minutes, 1, 8, 24, 48, 72, 96, 120, 144 and 168 hours following administration of the liposome formulation. The animals were then sacrificed by cervical dislocation and their heart, liver, kidneys and spleen were harvested. Each organ was thoroughly washed in phosphate buffer saline (PBS, pH = 7.4) and then frozen at -80°C. The plasma was isolated by centrifugation of the blood samples at 3000 g for 10 minutes. Iohexol and gadoteridol were extracted from the plasma and tissue samples using 10% perchloric acid (4-fold excess volume). Plasma and tissue concentrations of iohexol

47 were determined using a high performance liquid chromatography instrument (HPLC, PerkinElmer Series 200) equipped with a C18 Xterra reverse-phase column with ρaminobenzoic acid as the internal standard. The mobile phase for plasma samples was 90% methanol and 10% 100mM acetic acid buffer at a pH of 4.10. The mobile phase for tissue samples was composed of 92% methanol and 8% 100mM acetic acid buffer at a pH of 4.10. The flow rate was 0.9 mL/min and UV detection was performed at 245 nm to measure the concentration of iohexol. The plasma and tissue concentrations of gadoteridol were determined using ICP-AES [61]. The data obtained from the pharmacokinetics study was used to determine the main pharmacokinetic parameters for iohexol and gadoteridol when administered as free agents or agents encapsulated within liposomes. For the free agents, a two-compartment model was used to determine the distribution constant (Kd or α) and the elimination constant (Ke or β). The distribution half-life (t1/2α ) was then calculated using the equation: t1/2α = ln(2)/Kd, while the elimination half-life (t1/2β) was calculated using the equation: t1/2β = ln(2)/Ke. For the liposome-encapsulated agents, the Ke value was determined by fitting the plasma concentration versus time curve (each data point represents the mean of three distinct animals) with a one-compartment model. The vascular circulation half-life (t1/2) was then calculated using the following equation: t1/2 = ln(2)/Ke. The area under the plasma concentration versus time curve (AUC) was calculated using the trapezoid rule. The plasma clearance CL and the volume of distribution Vd were determined using Equations 3.1 and 3.2, respectively, as shown below. CL =

Dose AUC ⋅ BodyWeight

(Equation 3.1)

48

Vd =

CL Ke

(Equation 3.2)

Due to the inadequate resolution of the clinical CT and 1.5 T MR (and head coil) systems for imaging mouse vasculature, a larger animal model (rabbit) was employed for the following imaging studies.

CT and MR imaging of animal subjects The in vivo imaging study was performed under a protocol approved by the University Health Network Animal Care and Use Committee. Healthy female New Zealand White rabbits (2.5-3 kg) were anaesthetized with an intramuscular injection of either a ketamine and xylazine mixture or acepromazine. A slow bolus injection (0.5 mL/second) of 20 mL of the liposomal contrast agent formulation was then administered to the marginal ear vein catheter. Each rabbit received 730 mg/kg iohexol (equivalent to 340 mg/kg of iodine) and 69 mg/kg gadoteridol (equivalent to 19 mg/kg of gadolinium) co-encapsulated within the liposomes. 2% isoflurane vapor was given by inhalation throughout the study. Images of the rabbits were acquired pre and post-administration of the liposome formulation in CT (GE Discovery ST, General Electric Medical Systems, Milwaukee, WI, USA) and MR (GE Signa TwinSpeed MR scanner, General Electric Medical Systems, Milwaukee, WI, USA). The rabbits were CT scanned (120 kVp, 200 mA, a voxel size of 0.43 x 0.43 x 0.625 mm3, and a FOV of 220 x 220 x 400 mm3) at 10 and 60 minutes as well as 24, 48, 72, 96, 120 and 168 hours following administration of the liposome formulation. The rabbits were MR scanned (3D FSPGR sequence with a TR of 9.8 ms, a TE of 4.3 ms, a flip angle of 15°, a voxel size of 0.86 x 0.86 x 1.5 mm3 over a FOV of 220 x 220 x 228 mm3, and an image matrix of 256 x 256) at 30 and 90 minutes as well as 24, 48, 72, 96, 120 and 168 hours post-administration of

49 the formulation. The mean attenuation values in Hounsfield units (HU) in CT and the relative signal intensities (SI) in MR were measured in the aorta with circular regions of interest of over a cross sectional area of ~ 9 mm2 in a single axial image. For visualization purposes, 3D maximum intensity projection (MIP) images were generated using eFilm Workstation (Merge eFilm, Milwaukee, WI, USA). The same window and level were used for the pre and post-contrast images. In addition, for the correlation study, 1.5 mL of blood was collected from the ear vein of the same rabbits at the following time points: 5 minutes, 24, 48, 72, 96, 120 and 168 hours.

Acute toxicity studies and corresponding statistical analysis Female Balb-C mice (8-12 weeks, 18-20 g) were randomly divided into three groups as follows: mice receiving no formulation, mice receiving empty liposomes (530 mg/kg of lipid); mice receiving iohexol (650 mg/kg, equivalent to 300 mg/kg iodine) and gadoteridol (53 mg/kg, equivalent to 15 mg/kg gadolinium) co-encapsulated within liposomes (530 mg/kg lipid). Seven days later blood samples (0.5-1mL) were drawn by cardiac puncture and sent to Vita-Tech (Markham, Ontario, Canada) for haematological and biochemical analysis. The analysis included determination of number of white and red blood cells (WBC and RBC), platelets, and measurement of hematocrit, hemoglobin, serum creatinine, alkaline phosphatase (ALP), alanine transaminase (ALT) and aspartate transaminase (AST) concentrations. Statistical comparisons of the acute toxicity values were performed using the student t-test [174]. Computations were performed in Microsoft Excel. P-values greater than 0.05 were considered to be statistically insignificant.

50

3.4.

Results

Preparation and characterization of the multimodal liposome formulation The preparation, physico-chemical characterization and in vitro optimization of this multimodal liposome formulation have been described in detail elsewhere [61]. The average diameter of the liposomes in each preparation, as measured by DLS, was found to range between 70-85 nm. Each formulation contained an iodine to lipid weight ratio of 1:1.8 and gadolinium to lipid weight ratio of 1:35.7. The iodine to gadolinium ratio employed in this formulation was selected from consideration of in vitro imaging studies in phantoms, which evaluated the sensitivity of each imaging modality to detect the presence of contrast material within the formulation [61].

Pharmacokinetics and biodistribution studies in healthy mice The pharmacokinetics and organ distribution profiles of the co-encapsulated contrast agents, iohexol and gadoteridol, were evaluated in healthy female Balb-C mice as a means to assess the in vivo stability of this liposome formulation. Figure 3.1 includes the 7-day pharmacokinetics profiles for iohexol and gadoteridol, following i.v. administration in the DPPC/CHOL/PEG2000DSPE liposomes, as well as the 3-hour pharmacokinetics profiles for free iohexol and gadoteridol. The pharmacokinetics profiles for the agents encapsulated in liposomes were fit using a one-compartment model and the main pharmacokinetics parameters were calculated as listed in Table 3.1. The circulation half-lives for the agents were found to be 18.4 ± 2.4 hours for liposome encapsulated iohexol and 18.1 ± 5.1 hours for liposome encapsulated gadoteridol. The pharmacokinetics profiles for the free agents were fit using a two-compartment model. The distribution (α phase) half-life for free iohexol was

51 12.3 ± 0.5 minutes and for free gadoteridol it was 7.6 ± 0.9 minutes, while the elimination (β phase) half-lives were 3.0 ± 0.9 hours for free iohexol and 3.0 ± 1.3 hours for free gadoteridol. The values obtained for the half-lives of the free agents are in agreement with previously published results [175, 176]. The extended and similar circulation half-lives obtained for iohexol and gadoteridol when administered in this liposome formulation suggest that these agents remain co-encapsulated within the formulation in vivo. Figure 3.2 includes biodistribution profiles for the liposome-encapsulated agents in the heart, liver, kidney and spleen over a 7-day period. Similar distribution and clearance behavior were seen in the heart and liver for iohexol and gadoteridol. While an enhanced elimination of iohexol was observed in the kidney and the spleen compared to gadoteridol.

In vivo CT and MR imaging in healthy rabbits Imaging studies were performed on rabbits with a clinical CT scanner and a clinical MR scanner with a head coil.

As shown in Figure 3.3, the same rabbit was imaged

sequentially in CT and MR for a period of 7 days at selected time points both prior to and following administration of the liposome formulation. The clear post-contrast visualization of the rabbit heart, liver and spleen, as well as the transient visualization of the kidneys, is in agreement with the presence of the liposomal iohexol and gadoteridol detected in the same organs in mice (Figure 3.2). At each time point a 1 mL sample of blood was also collected from the rabbit and the plasma concentrations of agents present were quantified using HPLC and ICP-AES analysis. A region of interest of 2 mm in diameter in the rabbit aorta was identified and the signal

52 changes were measured in CT and MR and compared to the concentration values for iodine and gadolinium as determined by analysis of the plasma samples.

1000 100 10

Plasma agent concentration (µg/mL)

Plasma agent concentration (µg/mL)

10000

10000

1 0.1

1000 100 10 1

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Time (h)

0.01 0

25

50

75

100

125

150

175

200

Time (h) Figure 3.1 Pharmacokinetics of free iohexol (
), free gadoteridol (), liposomal iohexol () and liposomal gadoteridol () in healthy female Balb-C mice (n=3). The 2-week-old mice (18-23 g) were i.v. administered free iohexol and free gadoteridol diluted in HBS or liposome encapsulated iohexol and gadoteridol containing 650 mg/kg of iohexol (equivalent to 300 mg/kg iodine) and 60 mg/kg gadoteridol (equivalent to 17 mg/kg of gadolinium). Plasma was sampled at the indicated time points and analyzed using HPLC for iohexol and ICP-AES for gadoteridol. Data are represented as the mean ± standard deviation.

53

Ke 2

r t1/2 (h) AUC (µg*h/mL) CL (mL/h/g) Vd (mL/g)

Iohexol

Gadoteridol

0.0377

0.0383

0.975 18.4 5910000 0.00219 0.0580

0.980 18.1 582000 0.00206 0.0538

Table 3.1 Pharmacokinetic parameters for iohexol and gadoteridol when administered in a liposome formulation to female Balb-C mice. Abbreviations: Ke is the elimination constant; r2 is the coefficient of determination for this fit (every point used for the fit is the mean value obtained from 3 distinct animals); t1/2 is the vascular circulation half-life; AUC is the area under the concentration versus time curve in plasma; CL is the total plasma clearance and Vd is the volume of distribution per unit mass.

54 (a)

µg agent / g heart

10000

1000

100

10

1 0

25

50

75

100 125 150 175 200 Time (h)

(b)

µg agent / g liver

10000

1000

100

10

1 0

25

50

75

100 125 150 175 200 Time (h)

55 (c)

µg agent / g kidney

10000

1000

100

10

1 0

25

50

75

100 125 150 175 200 Time (h)

(d)

µg agent / g spleen

10000

1000

100

10

1 0

25

50

75

100 125 150 175 200 Time (h)

Figure 3.2 Biodistribution of iohexol (
) and gadoteridol () when administered in a liposome formulation to female Balb-C mice. The animals were sacrificed at specific times and a) heart, b) liver, c) kidneys, d) spleen samples were analyzed to determine levels of iohexol and gadoteridol. Each data point represents the mean of three distinct animals ± standard deviation.

Figure 3.3 Three-dimensional maximum intensity projection images (anterior view) of a healthy New Zealand White rabbit (3kg) obtained in CT (120 kV, 200mA) and MR (3D FSPGR sequence, TR/TE=9.8/4.3) prior to and following i.v. administration (as indicated) of the liposome formulation of iohexol and gadoteridol. The same window and level were used for pre- and post-injection images. Note the visual contrast changes in the heart (H), aorta (A), vena cava (V), carotid artery (C), kidney (K) and spleen (S).

56

57 The percent signal increases in CT and MR were calculated using equations 3 and 4.

% HU increase =

%SI increase =

(HU

(SI

t

− HU t o

HU t o t

− SI t o

SI t o

) ⋅ 100

) ⋅100

(Equation 3)

(Equation 4)

Figure 3.4 includes a plot of the signal changes in CT and MR versus the measured value of the concentration of each respective agent in plasma. A linear correlation (r2=0.997) was obtained for the %HUincrease measured in CT and the concentration of iodine in the plasma (Ciodine). In contrast, an exponential relationship was obtained for the %SIincrease measured in MR and the plasma concentrations of gadolinium (Cgadolinium). This is a result of the established non-linear relationship between MR signal intensity and gadolinium concentration [177]. The successful correlation of the signal changes measured using the imaging systems and the actual concentration of contrast agents detected in the biological samples indicates that this liposome formulation may be suitable for quantitative imaging applications, as well as non-invasive and quantitative CT and MR tracking of these nanosized vehicles in vivo.

Preliminary evaluation of acute toxicity Figure 3.5 summarizes the results obtained from the hematological and biochemical analysis of plasma samples obtained one week following administration of both empty liposomes and the liposome formulation of the CT and MR contrast agents. As shown, there were no statistically significant changes (for p=0.05) in the levels of red and white blood cells, hemoglobin, hematocrit, serum creatinine, and various liver enzymes (ALP, ALT and AST), 7 days following administration of the multimodal liposomes in comparison to

58 animals receiving no treatment or those that received the empty liposomes. This analysis provides a preliminary indication of the lack of toxicity and biocompatibility of this formulation.

(a) 400

HU increase (%)

350 300 250 200 150 100 %HUincrease = 0.183*Ciodine- 9.65

50

2

R =0.997

0 0

300

600

900

1200 1500 1800 2100

Ciodine (µg/mL) (b) 300

SI increase (%)

250 200 150 -b*Cgadolinium

100

%SIincrease= a*(1 - e

50

a = 326.2 + 31.8 b = 0.102 + 0.002

)

0 0

50

100

150

Cgadolinium (µg/mL)

200

250

59 (c)

% HUincrease in CT vs. Ciodine % SIincrease in MR vs. Cgadolinium

400

Signal increase (%)

350 300 250 200 150 100 50 0 0

300

600

900

1200 1500 1800 2100

Ciodine or Cgadolinium (µg/mL) Figure 3.4 Plots of the relative change in signal intensity pre- and post-administration of the multimodal liposomal agent (a) in CT versus the measured plasma iodine concentration, (b) in MR versus the measured plasma gadolinium concentration. The %HUincrease in CT was measured using circular regions of interest of 2 mm in diameter in the rabbit aorta and the plasma concentrations of iodine were determined by HPLC (
). The %SIincrease in MR was measured using circular regions of interest of 2 mm in diameter in the rabbit aorta and the plasma concentration of gadolinium was determined by ICP-AES (). (c) The two plots are combined in a single graph to illustrate the differential response of each modality to different concentrations of the respective contrast agent.

60

No-injection Empty liposomes (day 7) Iohexol and gadoteridol loaded liposomes (day 7) 1200 1000 800 600 400 200

AST (U/L)

ALT (U/L)

ALP (U/L)

Serum Creatinine (umol/L)

Platelets (x10e9/L)

Hematocrit (%)

Hemoglobin (g/L)

RBC (x10e9/L)

WBC (x10e9/L)

0

Figure 3.5 Summary of the hematological and biochemical evaluation of plasma samples obtained from female Balb-C mice (n=3) seven days following (1) no treatment, (2) administration of empty liposomes, or (3) administration of liposomes containing both iohexol and gadoteridol. Abbreviations: white blood cell (WBC), red blood cell (RBC), alkaline phosphatase (ALP), alanine transaminase (ALT) and aspartate transaminase (AST). Data are represented as the mean ± standard deviation. For all parameters, the differences between the 3 groups are found to be statistically insignificant using the student t-test (all pvalues were greater than 0.05).

61

3.5.

Discussion The in vivo stability of this liposome formulation was confirmed by evaluation of the

pharmacokinetics (PK) and biodistribution of the co-encapsulated agents, iohexol and gadoteridol, in healthy mice at various time points following intravenous administration. As shown in Table 3.1, the circulation half-lives for iohexol and gadoteridol were 18.4 ± 2.4 hours and 18.1 ± 5.1 hours, respectively, when administered in this liposome formulation. When the free agents were administered at the same dose, the distribution (α phase) half-life for the free iohexol was 12.3 ± 0.5 minutes and 7.6 ± 0.9 minutes for the free gadoteridol; while, the elimination (β phase) half-lives were 3.0 ± 0.9 hours for free iohexol and 3.0 ± 1.3 hours

for

free

gadoteridol.

Thus,

formulation

of

these

agents

in

the

DPPC/CHOL/PEG2000DSPE liposomes significantly increases their circulation half-lives. Though efforts were not put forward to distinguish between the encapsulated and released iohexol or gadoteridol, the similar behavior of iohexol and gadoteridol in terms of accumulation and clearance as detected in the blood, heart and liver strongly suggests that these agents are still co-encapsulated within the internal aqueous volume of the liposomes at the time of measurement. However, iohexol shows an enhanced elimination in both the kidney and the spleen compared to gadoteridol. This may be attributed to the different mechanisms associated with the clearance and metabolism of the individual contrast agents in the kidneys [178, 179] and the spleen. Studies have shown that following intravenous administration approximately 95% of the free agents are cleared though the glomerular filtration process in the kidneys [175, 180]. Consequently no study has yet been conducted to investigate the clearance and metabolism of iohexol and gadoteridol in the spleen. The alteration in the biodistribution of these agents due to administration in the liposome

62 formulation has now prompted a separate study to investigate the clearance of these agents from the spleen. The contrast enhancement seen in CT and MR, as shown in Figure 3.3, is due to the increased iodine and gadolinium content in the visually enhanced locations. Unlike radionuclide and optical imaging which employ radionuclide tracers and optical labels, CT and MR contrast agents such as iohexol and gadoteridol do not decay or bleach over time. Hence CT and MR are two imaging methods suitable for multi-session longitudinal studies, especially those requiring long imaging sequences. One of the advantages of this dual CT and MR contrast agent system is that the in vivo agent concentrations may be monitored over a wider range. Figure 3.4 shows the ability of MR to estimate in vivo gadolinium concentrations ranging from 10 µg/mL to 200 µg/mL, while CT can estimate iodine concentrations from 100 µg/mL to 2000 µg/mL. The lower detection limit presented here corresponds to the iodine or gadolinium concentration needed to generate a signal differential in CT or MR that is greater than the highest noise level. The two imaging modalities may, therefore, detect in vivo liposome concentrations that are one thousand fold lower (~ 1011 liposomes/mL) than the original formulation administered (~ 1014 liposomes/mL). In this way, the dual liposome-based CT and MR contrast agent allows for measurement over a broader concentration range, and also takes advantage of the strengths of each imaging modality. For example, CT provides contrast of the bony structures with high spatial and temporal resolution, while MR allows for better visualization of the soft tissues [76, 77, 181]. The colloidal size of this multimodal liposomal agent makes it a good intravascular agent (Figure 3.3) that may be able to provide reliable estimation of vascular volume. Currently available small molecular weight contrast agents exhibit two-compartment

63 pharmacokinetics, quickly leaking from the blood vessels into the tissue interstitium. Thus, they require complex physiological modeling as well as fast imaging sequences in order to measure their first pass enhancement in studies involving deconvolution of blood vessel permeability and vascular volume in angiogenic tumors [182-184]. The successful development of this liposomal agent with prolonged intravascular residence time may be of assistance in obtaining more accurate perfusion and permeability measurements in healthy and diseased tissues, as well as information on physiological processes that occur over a longer time course. Following characterization of the in vivo stability and behavior of this liposome-based system it also became evident that this system may be used as a tool to address unanswered questions that remain surrounding the in vivo fate of passively and actively targeted nanocarriers. The clear advantages to the use of imaging methods, over conventional whole organ digestion methods, to map liposome distribution in vivo is the non-invasive nature of this approach and the ability to also obtain sub-organ or sub-tissue distribution patterns up to the spatial resolution limit of the imaging system. Specifically in this study, the voxel size achieved was 0.43 x 0.43 x 0.625 mm3 in CT and 0.86 x 0.86 x 1.5 mm3 in MR. Potential applications of this CT and MR system include non-invasive assessment of tumor accumulation and distribution of passively and actively targeted liposomes in pre-clinical and clinical settings, development of correlations between tumor penetration of liposomes and the state of tumor vasculature [185]. In addition, this multimodal liposome system may be used to assess the performance or behavior of liposomes following administration of different therapies (i.e. anti-angiogenic therapies, radiotherapy and/or chemotherapy), which may ultimately aid in the optimization of the sequence and dosing of combined therapies.

64

3.6.

Acknowledgements This work is funded in-part by a CIHR Operating Grant and a CIHR Proof of

Principle Grant to D.A. Jaffray and C. Allen, the Premier’s Research Excellence Award, the Fidani Chair in Radiation Physics and the Grange Advanced Simulation Initiative. J. Zheng is grateful for the Excellence in Radiation Research for the 21st Century Training Fellowship and the Mitchell Scholarship. The authors would like to thank the UHN animal care staff for their assistance.

Chapter 4. Quantitative CT Imaging of the Spatial and Temporal Distribution of Liposomes in a Rabbit Tumor Model

65

66

4.1.

Foreword The previous two chapters demonstrated the stability of the liposome system in vitro

and in vivo. This chapter explores the use of volumetric regions of interests in CT to measure the changes in iodine concentrations in normal and diseased tissues over time. The sensitivity (i.e. of minimum amount of detectable iodine) of this method was assessed as a function of the size of the volume of analysis. The method was then applied to non-invasively characterize the biodistribution and kinetics of liposomes in a VX2 carcinoma rabbit model over a 14-day period. The sub-millimeter spatial resolution of CT allowed for visualization of the heterogeneity of contrast enhancement within tumors, allowing quantification of the intratumoral volume of distribution of liposomes over time. The following chapter has been published as: Zheng J, Jaffray DA, Allen C. Quantitative CT Imaging of the Spatial and Temporal Distribution of Liposomes in a Rabbit Tumor Model. Molecular Pharmaceutics, Volume 6, Number 2, Pages 571-580. March 2009. It has been reproduced with permission from the American Chemical Society. Copyright 2009 American Chemical Society.

4.2.

Introduction Characterization of the pharmacokinetics and biodistribution of novel imaging and

therapeutic agents is critical for understanding their potential performance and effectiveness

in vivo [186-189]. In recent years, developments in imaging techniques have provided new tools for non-invasive visualization of the spatial and temporal distribution of these agents through labeling with fluorescent, radioactive, radioopaque or paramagnetic molecules and

67 assessment using optical, single photon computed tomography (SPECT), positron emission tomography (PET), CT and magnetic resonance imaging (MRI), respectively [190-195]. The non-invasive nature of image-based assessments allows for repeated in vivo and in situ data acquisition from the same subject over multiple time points, thereby reducing the required number of animals while increasing the accuracy of measurements. Furthermore, if a high resolution imaging technique is employed, the intra-organ and tissue distribution of the agent can be resolved. As a result, imaging has become utilized increasingly for biodistribution investigations; however, appropriate extraction of quantitative data from images still remains a challenge. Nano-sized lipid nanoparticles such as liposomes have been widely employed as delivery vehicles for a range of molecules such as drugs and contrast agents. Their size and surface properties have proven to be critical for passive accumulation in tumors and sites of inflammation through the enhanced permeation and retention (EPR) effect [21, 196-198]. Their in vivo distribution has been assessed by numerous research groups in a variety of healthy and disease-bearing animal models as well as in patients using both traditional tissue extraction [199, 200] and nuclear imaging techniques [46, 201]. SPECT imaging techniques rely on radioisotope labeling of liposomes. As a result, the imaging time window is limited by the physical half-life of gamma photon emitting isotopes such as 99mTc (t1/2 = 6 h) or In111 (t1/2 = 67 h), and increased image acquisition time is necessary to compensate for radioisotope decay. CT contrast agents such as iodine and barium are non-radioactive, have high atomic numbers and provide high x-ray attenuation. The employment of a CT-based assessment is therefore suitable for investigations involving long-circulating nanoparticle systems, as well as for monitoring slow physiological processes such as the passive

68 accumulation of nano-carriers in tumors via the EPR phenomenon. Furthermore, volumetric CT imaging allows for extremely fast data acquisition in sub-millimeter isotropic voxels. When combined with 3D image analysis tools, volumetric quantification of signal profiles within an organ or tissue of interest is possible. This enables the performance of whole body mass balance calculations and quantification of intra-organ heterogeneity. In addition, CT is currently the fastest and most widespread whole body volumetric imaging modality, which makes it very attractive for high throughput biodistribution investigations. The goal of the current research is to longitudinally quantify the presence of iohexol and gadoteridol-containing liposomes in the various body compartment volumes, as well as to visualize the heterogeneity of liposome distribution within a tumor using volumetric highresolution CT imaging.

4.3.

Experimental Section

Materials The lipid components of the liposome bilayer 1,2-Dipalmitoyl-sn-Glycero-3Phosphocholine (DPPC, M.W. 734) and 1,2-Distearoyl-sn-Glycero-3-PhosphoethanolamineN-[Poly(ethylene glycol)2000] (PEG2000DSPE, M.W. 2774) were purchased from Genzyme Pharmaceuticals (Cambridge, MA, USA); cholesterol (CH, M.W. 387) was purchased from Northern Lipids Inc. (Vancouver, British Columbia, Canada).

The CT contrast agent

Omnipaque (Nycomed Imaging AS, Oslo, Norway) has an iodine concentration of 300 mg/mL and consists of the non-ionic, iodinated molecule iohexol (N, N'-Bis(2,3dihydroxypropyl)-5-[N-(2,3-dihydroxypropyl)-acetamido]-2,4,6-triiodo-isophthalamide, M.W. 821.14, 3 iodine atoms per molecule) dissolved in an aqueous solution with

69 tromethamine and edentate calcium disodium. The non-ionic, gadolinium complex gadoteridol

(10-(2-hydroxy-propyl)-1,4,7,10-tetraazacyclododecane-1,4,7-triacetic

acid,

M.W. 558.7, 1 gadolinium atom per complex, ProHance by Bracco Diagnostics Inc., Princeton, NJ, USA) dissolved in an aqueous solution with calteridol calcium and tromethamine (gadolinium concentration of 78.6 mg/mL) was also encapsulated in the liposomes.

Preparation and Characterization of Liposome Formulations The liposome composition (DPPC, cholesterol and PEG2000DSPE in 55:40:5 percent mole ratios) and preparation method were described in detail in previous publications [61, 62]. The mean diameter of the final liposome sample was measured by dynamic light scattering (DLS) analysis of dilute solutions using a DynaPro DLS instrument (Protein Solutions, Charlottesville, VA, USA) at 25°C. The final concentration of iohexol was determined using a UV assay with detection at a wavelength of 245 nm (Cary 50 UV/VIS Spectrophotometer, Varian Inc, CA, USA). The final concentration of gadoteridol was determined using an assay based on inductively coupled plasma atomic emission spectrometry (ICP-AES Optima 3000DV, Perkin Elmer, MA, USA).

CT Imaging of Tumor Bearing Rabbits The following in vivo imaging study was performed under a protocol approved by the University Health Network Animal Care and Use Committee. Five healthy male New Zealand White rabbits (2.8-3.2 kg) were inoculated with approximately 400 µL of VX2 carcinoma cells which were obtained from two propagation rabbits. The tumour cells were

70 injected intramuscularly into the animals’ left lateral quadriceps. The contrast-enhanced imaging studies were performed seven to ten days after the tumour inoculation procedure. Specifically, each rabbit was intubated and kept under anaesthesia with a mixture of isoflurane and oxygen via inhalation throughout each imaging session. A slow bolus injection (0.5 mL/second) of approximately 15 mL of the liposomal contrast agent formulation was then administered via the marginal ear vein catheter. Each rabbit received 595 mg/kg of iohexol (equivalent to 276 mg/kg of iodine, corresponding to approximately half of the iodine dose/body weight typically administered to patients in a bolus form) and 40 mg/kg gadoteridol (equivalent to 11 mg/kg of gadolinium) co-encapsulated within liposomes of 80.2 ± 3.4 nm in diameter and 6.2 ± 4.3 % in polydispersity. CT Images (GE Discovery ST, General Electric Medical Systems, Milwaukee, WI, USA) of the rabbits were acquired pre and post-administration of the liposome formulation at 30 minutes as well as 1, 2, 3, 5, 7, 10 and 14 days following administration of the liposome formulation using the following imaging parameters: 80 kVp, 200 mA, a voxel size of 0.43 x 0.43 x 0.625 mm3, and a FOV of 220 x 220 x 400 mm3. The nominal x-ray dose for one whole body scan estimated by the scanner’s CT dose index is 15 mGy. In addition, urine and feces samples were collected on a daily basis and analysed for iodine content using neutron activation analysis (Becquerel Laboratories, ON, Canada) for assessment of liposome clearance route(s) and kinetics.

Volumetric Analysis of the CT Data Sets Semi-automated contouring using MicroView v2.2 allowed for generation of threedimensional volumes of interest consisting of the left and right kidneys, spleen, liver, tumor and the contra lateral muscle. The mean HU value for a given organ or tissue was calculated

71 by averaging the signal of all voxels within the contoured volume. A voxel number versus CT signal profile was also generated for each volume at each imaging time point. All profiles were fit to a Gaussian curve with R2 values greater than 0.90. The sigma of the signal profiles is a compounded result of the heterogeneity of the biological system and uncertainty in the measurement method. For the bulk volume analysis, an assumption was made that the organs and tissues of interest were homogeneous. The uncertainty in the measurement method was obtained by CT imaging a large water phantom and measuring the standard deviation of the signal within volumes of similar size as the organs and tissues of analysis (ranging from 0.2 to 40.9 cm3). The Welch’s t-test was used to calculate the degree of significance between the CT signals within the same organ over the different time points. The difference between the mean Hounsfield unit (HU) measured at time t post-liposome injection and the mean HU measured pre-liposome administration at time t=0 (Equation 4.1). ∆meanHU = meanHU t − meanHU t = 0

4.4.

(Equation 4.1)

Results

Liposome Accumulation and Clearance Kinetics in Organs and Tissues of Interest An axial image representing each organ and tissue of interest is shown in Figure 4.1a. In Figure 4.1b, the organ volumes that were contoured and analyzed are illustrated in yellow with respect to their locations within the rabbit body. The liposome accumulation and clearance kinetics profiles in each organ (left and right kidneys, liver, spleen and tumor) of each of the five animals are shown in Figure 4.1c as ∆meanHU (Equation 4.1). In all volumes of interest and in all five animals, a sharp increase in the mean HU value was seen immediately following the administration of the liposome agent (30 minutes

72 post-injection) as a result of the systemic distribution of liposomes in the blood stream. While the mean signal intensities measured in the healthy organs (kidneys, liver and spleen) decreased over time, significant contrast enhancement of the tumor volume is not observed until 24 hours post-injection, and it is sustained up to 10 days following a single administration of the liposomes (Figure 4.2a). As a control, analysis was also performed on the muscle volume located on the contra-lateral thigh. The highest tumor-to-muscle iodine concentration ratio of 11.9 ± 6.0 was detected at 7 days post-injection (Figure 4.2b). However, the highest liposome accumulation (915 µg/cm3 of iodine) at the tumor site occurred at 48 h following administration. The linear relationship between differential CT attenuation values and iodine concentration was determined in a separate phantom study. It was measured that every 1 mg/mL of iodine and 0.05 mg/mL of gadolinium encapsulated in liposomes provided a differential signal increase of 38.04 ± 0.64 HU in CT when operated at 80 kVp and 200 mA. The coefficient of determination R2 for the linear regression was 0.996.

73

Figure 4.1 Visual illustration of (a) axial CT slices of the rabbit kidneys, liver, spleen and tumor acquired at 48 h post-injection. These images are acquired at sub-millimeter resolution and they demonstrate potential for quantification of intra-organ heterogeneity. In this particular study, bulk organ analysis was performed on (b) the contoured organ/tissue volumes (in yellow). (c) The differential mean HU measured in each volume of interest (with respect to the pre-injection data set) at selected time points. Each profile represents the values obtained for a given rabbit over 14 days.

74 (a) 5000 Blood Left Kidney Right Kidney Spleen Liver Tumor Muscle

3

Iodine concentration (µg/cm tissue)

4500 4000 3500 3000 2500 2000 1500 1000 500 0 0

50

100

150

200

Time (h)

250

300

350

75

Tumor to Muscle Iodine Concentration Ratio

(b) 20 18 16 14 12 10 8 6 4 2 0 0

50

100

150

200

250

300

350

Time (h)

Figure 4.2 (a) Liposome biodistribution profiles in the various organs and tissues of interest as measured using CT-based detection of the co-encapsulated iohexol and gadoteridol. The encapsulated iodine to gadolinium weight ratio is 20 to 1. At day 14, the blood, spleen, tumor and muscle showed statistically significant accumulation of liposomes (p < 0.001), while the mean signal measured in the kidneys and the liver were not statistically significant compared to the mean signal of the same organs pre-liposome administration. (b) Time-dependent tumor-to-muscle ratio of iodine concentration. The highest ratio occurs at 7 days postliposome injection, which coincides with the highest liposome accumulation detected in the tumor. Each data point represents the mean ± standard deviation for five animals.

76

Time-Dependent Biodistribution as a Function of Injected Dose Conventional biodistribution studies and nuclear medicine imaging techniques report the amount of drug or agent extracted or the total radioactivity measured from a given organ or tissue, respectively, as a percentage of the injected dose (%ID) and as percentage of the injected dose per weight (%ID/g or %ID/kg) of the organ or tissue of interest. This data is measured from the CT data set using an anatomically accurate volumetric organ-based analysis technique. Table 4.1 displays the percentage of injected iodine per organ and the percentage of injected iodine per volume (cm3) of tissue for blood, kidneys, liver, spleen and tumor. Our findings are in agreement with a study performed by Harrington et al. [46] in cancer patients who had been administered

111

In-DTPA labeled Doxil® liposomes and

imaged with SPECT. It is important to note that there are physiological differences between rabbits and human. The lipid composition of the liposomes employed for this study (DPPC : cholesterol : PEG2000DSPE at 55:40:5 mol%) is fairly similar to the Doxil® liposome formulation (HSPC : cholesterol : PEG2000DSPE at 56:39:5 mol%). Previously, our group has shown that the biodistribution profile of this liposome formulation in healthy mice matched the profiles reported by other groups who used either Doxil® liposomes or other liposome formulations that closely matched the composition of Doxil® [62, 202]. Table 4.1 shows that the %ID of iodine measured in the rabbits’ vascular compartment using CT is 89.8 ± 19.5% at 30 minutes post-injection and decreases to 51.4 ± 4.2% at 48 hours and then to 10.9 ± 3.3% at day 10. In comparison, Harrington et al. reported that 95.0 ± 11.8% of the %ID of 111

In (i.e. 111In-Doxil) remained in patients’ blood 30 minutes post-liposome administration,

55.5 ± 9.3% at 48 hours and 4.9 ± 5.1% at day 10. The blood pharmacokinetics profile from each rabbit was fitted to a one-compartment model and the mean vascular half-life t1/2 was

77 calculated to be 63.6 ± 5.8 h. The R2 values indicative of the goodness of fit ranged between 0.90 and 0.97 across the study population. In addition, the liposome accumulation in each rabbit kidney was measured to be between 1.9 ± 0.5% (0.5 h) and 0.1 ± 0.1% (240 h), in agreement with the values reported by Harrington et al. of 1.6 ± 0.8% (0.5 h) and 0.7 ± 0.4% (240 h) in patients. As well, the tumor accumulation in rabbits was measured to be 0.9 ± 0.3% in this study at 72 hours postinjection, which falls within the range of 0.3 - 2.6 %ID reported for patients having different types of tumor burden and treated with Doxil® liposomes. However, the %ID of liposomes measured in the rabbit liver and spleen, was about four times lower than the values reported by Harrington et al. in patients. In fact, a greater degree of cumulative excretion via the urinary route was observed in the current study (27.0 ± 15.7 %ID) compared to the 18.3 ± 6.9 %ID reported by Harrington et al. over the first 4 days post-injection. Stool samples collected from all five animals over the entire study period revealed a cumulative 7.3 ± 9.8 %ID of iodine excreted at day 4 and a cumulative 12.6 ± 15.1 %ID excreted at day 14. Overall, with the combination of the CT volume analysis method for blood, kidneys, liver, spleen and tumor, and iodine detection in urine and feces, it was possible to account for 100.8 ± 22.0% of the total injected dose of iodine at 30 minutes post liposome administration, 80.8 ± 12.2% of the total injected dose at 72 h post-injection, and 58.5 ± 9.4% of the total injected dose at 14 days post-administration. The remaining amount of iodine may be non-specifically distributed in other body compartments that were not included in this analysis such as the skin, muscle, fat or the interstitial fluid space.

78

Blood

Kidneys

Liver

Time (h)

Spleen

Tumor

Urine

Feces

% ID

0.5

89.8 ± 19.5%

3.9 ± 0.6%

5.9 ± 2.7%

0.8 ± 0.5%

0.4 ± 0.1%

-*

-*

24

59.3 ± 4.7%

1.7 ± 0.4%

3.9 ± 0.9%

1.0 ± 0.4%

0.7 ± 0.1%

18.8 ± 14.3%

1.3 ± 1.7%

48

51.4 ± 4.2%

1.3 ± 0.4%

4.4 ± 1.0%

1.0 ± 0.4%

0.9 ± 0.3%

22.9 ± 16.2%

4.2 ± 6.0%

72

44.2 ± 5.9%

1.0 ± 0.4%

4.3 ± 2.8%

0.7 ± 0.3%

0.9 ± 0.3%

24.2 ± 16.9%

5.5 ± 7.5%

120

33.1 ± 4.8%

0.8 ± 0.3%

3.0 ± 2.2%

0.5 ± 0.2%

1.1 ± 0.3%

29.4 ± 14.8%

8.5 ± 10.7%

168

21.1 ± 4.5%

0.5 ± 0.3%

2.2 ± 1.5%

0.3 ± 0.2%

1.1 ± 0.3%

35.3 ± 14.9%

10.3 ± 12.7%

240

10.9 ± 3.3%

0.3 ± 0.2%

1.2 ± 0.8%

0.2 ± 0.1%

1.0 ± 0.3%

39.5 ± 13.9%

11.8 ± 14.3%

336

2.1 ± 1.4%

-**

-**

0.1 ± 0.0%

0.6 ± 0.3%

43.1 ± 12.3%

12.6 ± 15.1%

% ID / cm3

Time (h) 0.5

0.42 ± 0.04%

0.20 ± 0.08%

0.14 ± 0.05%

0.24 ± 0.10%

0.05 ± 0.05%

24

0.29 ± 0.06%

0.09 ± 0.02%

0.10 ± 0.03%

0.25 ± 0.08%

0.09 ± 0.02%

48

0.25 ± 0.06%

0.07 ± 0.02%

0.11 ± 0.03%

0.26 ± 0.07%

0.11 ± 0.01%

72

0.22 ± 0.06%

0.06 ± 0.02%

0.10 ± 0.04%

0.21 ± 0.06%

0.11 ± 0.01%

120

0.16 ± 0.05%

0.05 ± 0.02%

0.07 ± 0.03%

0.16 ± 0.04%

0.11 ± 0.01%

168

0.11 ± 0.04%

0.03 ± 0.02%

0.05 ± 0.02%

0.10 ± 0.03%

0.10 ± 0.01%

240

0.06 ± 0.03%

0.02 ± 0.02%

0.04 ± 0.01%

0.08 ± 0.02%

0.08 ± 0.02%

336

0.01 ± 0.01%

-**

-**

0.03 ± 0.01%

0.04 ± 0.02%

Table 4.1 Liposome biodistribution expressed as %ID and as %ID/cm3 of organ/tissue. The volume of the organs and tissues of interest were measured using the CT data set with the exception of the blood compartment. The blood volume for each animal was calculated as the injected dose divided by the y-intercept of the mono-exponential fit from the blood iodine concentration vs. time profile. The estimation rather than measurement of blood volume increases the uncertainty associated with the %ID and as %ID/cm3 values for blood pool. The high variance in the blood iodine content at 30 minutes post-injection is likely due to inaccuracies associated with the timing of the imaging session. The values for urine and feces are cumulative. Each table entry represents the mean ± standard deviation for five animals. * No urine or stool output. ** Iodine concentration in this organ for this time point was below the detection limit.

79

Sensitivity of CT in Detecting the Tissue Concentrations of Iodine-Labeled Liposomes CT imaging was performed on all animals pre and post contrast administration at selected time points. As stated in the methods, volumes of interest were generated from semiautomatic contours. The voxel number versus CT signal profiles generated were then Gaussian fitted with R2 greater than 0.95 for kidneys, liver, spleen and tumor and with R2 greater than 0.90 for blood. The Welch’s t-test was used between each pair of pre and postinjection volume sets to determine whether their signal profiles were statistically different (p < 0.001). The critical t value of 3.291 was then used to calculate the minimum differential HU needed for a given pre and post-injection data set pair to be determined to be statistically different. These values were then converted into µg/cm3 of iodine concentration representing the minimum amount of iodine that CT is able to detect in the different organs (Table 4.2). It is worth noting that all voxels within the liver, kidneys, spleen and tumor were used in the analysis in order to maximize statistical power. However, it was not possible to contour all voxels occupied by blood. As a result, the sensitivity in detecting iodine concentrations in blood can be improved by increasing the volume of analysis. In this case, due to the high iodine content circulating in the bloodstream, even at day 14, the blood iodine concentration (76.6 ± 39.5 µg/cm3) was above the limit of detection of the current method (11.4 µg/cm3).

Classification of Heterogeneity in the Intratumoral Distribution of Liposomes Once it had been established that the iodine detection sensitivity in the tumor is 1.8 µg/cm3 (equivalent to a differential mean HU increase of 0.11 ∆HU) for this particular study, it was possible to calculate and visualize the percentage of tumor volume that was occupied by iodinated liposomes. Figure 4.3 provides visual illustration of the accumulation and

80 clearance of the iodinated liposomes from the tumors of the five rabbits over the 14-day period. It may be noted that although the percentage of tumor volume occupancy by the liposomes is fairly consistent across the five data sets, the spatial distribution patterns differ significantly from animal to animal. The graph in Figure 4.4a shows the time dependent volume of distribution of liposomes in tumor as the fraction of the total tumor volume occupied. The percent occupancy peaked at 72 ± 5% 48 h post-injection.

Mean Sampling Volume (cm3)

Minimum Mean ∆HU for Significance (p < 0.001)

Iodine Detection Sensitivity (µg/cm3)

Blood

0.2

0.68

11.4

Kidney

9.0

0.10

1.6

Liver

40.9

0.05

0.8

Spleen

3.1

0.17

2.9

Tumor

11.4

0.11

1.8

Table 4.2 List of the mean organ and tissue sampling volume used for this study during the analysis of the CT data sets. For a given body compartment of a set mean volume, the minimum mean differential HU (∆HU) needed to detect statistically significant amounts of iodine was calculated using the Welch’s t-test.

81

Figure 4.3 (a)* Anterior views of 3D CT maximum intensity projections (MIP) of a representative VX2 carcinoma bearing male New Zealand White rabbit (3 kg) at 30 minutes,

82 24 and 48 hours post liposome administration. The arrows indicate the site of the tumor and the EPR effect is visualized through the gradual opacification of the tumor area resulting from the accumulation of the iohexol and gadoteridol containing liposomes. (b) The five quadrants represent data acquired from five distinct animals, with each quadrant displaying 3D maximum intensity projections of the segmented tumor volumes pre and up to 14-days post liposome injection. Note that although the percent volume of distribution (Vd) of liposomes in the tumor at the different time points is relatively similar, the intratumoral spatial distribution pattern greatly differs from animal to animal. In addition, the tumor growth process can also be monitored, visualized and effectively measured using CT (see Figure 4.4b). * Adapted from Zheng et al. [203] to illustrate the anatomical location of the segmented tumor volumes

83 (a) 1.0

Tumor Volume Fraction

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0

50

100

150

200

250

300

350

250

300

350

Time (h) (b) 40000

Rabbit 1 Rabbit 2 Rabbit 3 Rabbit 4 Rabbit 5

Tumor Volume (mm3)

35000 30000 25000 20000 15000 10000 5000 0 0

50

100

150

200

Time (h) Figure 4.4 (a) Representation of the tumor volume fraction occupied by liposomes over 14 days. Voxels with values greater than or equal to the µ + 2σ of the pre-injection tumor CT signal were considered contrast enhanced. Each data point represents the mean ± standard deviation for five animals. (b) Changes in tumor volume measured using CT in the five rabbits over 14 days.

84

4.5.

Discussion Recent developments in image-guided drug delivery are prompted by the belief that

an increased understanding of the biodistribution of drug carriers in individual patients will lead to improvements in personalized treatment design and delivery. Advances in nano-sized drug carriers have enabled improved delivery of anticancer drugs to the tumor site, by exploitation of the EPR phenomenon, while minimizing their non-specific distribution to healthy organs and tissues. EPR describes the mechanism by which macromolecules or nanoparticles are retained within healthy vasculature due to their colloidal size, but are able to accumulate in tumors due to the leakiness of the abnormal vasculature and lack of effective lymphatic drainage at these sites, in comparison to normal tissue [21, 196-198]. The addition of an imageable component to a drug carrier, in combination with the employment of volumetric imaging techniques, enables non-invasive visualization and quantification of the effectiveness of tumor targeting and sparing of healthy tissue. The successful translation of image-guided drug delivery to the clinical setting would permit timely adjustments of treatment regimens on a per patient basis and ultimately enable implementation of personalized medicines. In the current study, iohexol and gadoteridol were co-encapsulated within liposomes to enable CT and MR imaging of the nanoparticle biodistribution following administration. The physico-chemical characteristics and stability of this formulation have been described in detail elsewhere [61]. We also previously reported the pharmacokinetics profiles of iohexol and gadoteridol as free agents and liposome-encapsulated agents [62] to demonstrate the prolonged imaging window of the contrast agent encapsulated liposomes. Similar to other small molecules, un-encapsulated iohexol and gadoteridol exhibit rapid distribution and

85 clearance upon administration. Liposome encapsulation of the small molecular weight agents changed their profiles from biphasic to monophasic. In healthy Balb-C mice, the calculated

tα1/2 was 12.3 ± 0.5 minutes for iohexol and 7.6 ± 0.9 minutes for gadoteridol, the elimination (β phase) half-lives were 3.0 ± 0.9 hours for iohexol and 3.0 ± 1.3 hours for gadoteridol; whereas the vascular half-lives were 18.4 ± 2.4 and 18.1 ± 5.1 hours for iohexol and gadoteridol, respectively, when co-encapsulated in liposomes [62].

As a result, it is

reasonable to attribute the longitudinal signal increases observed in the blood and tissue compartments to the presence of the encapsulated agents. Once it is established that the images are indeed reporting the biodistribution of the nanoparticulate carriers, there remains a significant challenge in using imaging techniques to assess the in vivo performance of drug delivery systems. This challenge lies in the accurate extraction of quantitative data from the images. Firstly, it must be ensured that the given signal change measured in a given voxel corresponds to a consistent change in the concentration of the nanocarrier in the same volume. When using imaging modalities such as MR and CT, in which signal changes can occur as a result of endogenous changes in tissue properties, the benefit of having anatomical information comes with the challenge of extracting signal changes that are solely generated by the presence of contrast agents. For example, when conducting longitudinal studies lasting days to weeks in a tumor-bearing animal, the tumor morphology and physiology can change. The signal generation process of an anatomic MR data set relies on the tissue T1 and T2 relaxation parameters, which are known to be very sensitive to changes in tumor tissue properties such as local water concentration [204]. The sensitivity to these parameters, which have made MR so powerful for soft tissue and tumor characterization, increases the challenge of quantifying relaxivity

86 changes that are exclusively caused by shifts in the contrast agent concentration. The underlying signal generation process in CT is dependent on the x-ray attenuation profile of a tissue, which during disease progression undergoes a more gradual modification process than its relaxation properties. As a result, although the liposomes employed for this investigation co-encapsulate iohexol and gadoteridol and can be imaged using both CT and MR, CT was relied on exclusively for quantification. The second requirement for quantification of imaging data, in the case of a longitudinal study, is high confidence in defining corresponding volumes of interest for analysis across image sets acquired at different times. Although a voxel-based time-course analysis was not pursued here, both rigid and deformable image registration techniques have shown success, within a reasonable error range, in the identification of voxels across longitudinal data sets in organs and tissues that do not undergo significant anatomic changes over the course of the imaging study [205, 206]. However, these algorithms cannot be applied to tissues that significantly change either due to disease progression or treatment. For example, during this particular study (2 weeks), the rabbit tumor volumes tripled in size (Figure 4.4b). Due to the low confidence in accurately identifying the same set of intratumoral voxels over time, a decision was made to measure the mean signal changes over all voxels that make up the bulk tumor volume and classify groups of intratumoral voxels according to their HU values rather than their spatial distribution (Figure 4.4a). Lastly, accurate quantification of the in vivo nanocarrier biodistribution is best achieved with a 3D analysis method.

In situations when the nanocarrier is uniformly

distributed, 2D region of interest (ROI) analysis may be preferred due to its simplicity and speed, and the mean signal value obtained is representative of the entire volume of interest.

87 However, the nanocarrier distribution patterns are often heterogeneous or unknown, the employment of 2D ROIs for image analysis is vulnerable to variations in 2D ROI placement. In this study, 3D ROI-based analysis was used on all organs and tissues of interest, whether or not they were known to be homogeneous or heterogeneous, in order to provide unbiased and observer-independent measurements that reflected the signal intensities of all voxels within the volumes of interest over all time points. The variance in the signal distributions provides insight on the heterogeneity of the organ/tissue under investigation. These heterogeneities include variance in anatomy, physiology and differential uptake and clearance of the contrast agent. The use of CT is attractive in monitoring these heterogeneities longitudinally due to its geometric accuracy. An additional advantage in using this 3D whole organ/tissue volume-based analysis is that it increased the number of voxels that were sampled for each measurement. As a result, a higher statistical confidence was achieved when comparing the differences among the signal profiles measured in the same organ or tissue over time, allowing us to confidently report much lower iodine concentrations (i.e. 0.8 µg/cm3 for the liver, Table 4.2) compared to those published by other research groups that have employed 2D ROI analysis techniques [106, 151, 207]. This also demonstrates that there is an opportunity to decrease imaging dose while still satisfying the necessary level of iodine detection. In conclusion, the feasibility and effectiveness of quantitative CT-based measurements of the pharmacokinetics and biodistribution of a nanocarrier in vivo have been demonstrated. Longitudinal imaging was performed up to 14 days post-liposome administration. The vascular half-life of 63.6 ± 5.8 h for the liposomes enabled high accumulation at the tumor sites (915 µg of iodine /cm3 of tumor tissue at 48 h post-injection)

88 through exploitation of the EPR effect. Furthermore, although the intratumoral distribution of liposomes was highly heterogeneous and variable from animal to animal, the vehicle occupied the majority (72 ± 5%) of the total tumor volume at 48 h post-injection in the five rabbits. These observations support ongoing efforts in our research laboratory to develop robust metrics for spatial classification of the heterogeneous intratumoral distribution patterns of nanocarriers. These, in conjunction with molecular imaging tracers that report the different properties of the tumor micro-environment (i.e. FAZA-PET, FMISO-PET), will become a powerful tool set to elucidate the ability of drug carriers to deliver therapeutic agents to various intratumoral regions that have distinct sensitivities to treatment. In addition, increased employment of non-invasive, quantitative image-guided pharmacokinetics and biodistribution assessments in the development and pre-clinical testing of novel nanocarriers has the potential to greatly facilitate their clinical translation. Conversely, the adoption of imageable nanotherapeutics in the clinical setting, along with quantitative imaging systems and analysis tools, will positively impact therapy outcome through personalization of treatment delivery.

4.6.

Acknowledgements This work is funded in-part by CIHR and OICR research grants, the Fidani Chair in

Radiation Physics and the Grange Advanced Simulation Initiative. Jinzi Zheng is grateful for the CIHR Canada Graduate Scholarship. The authors would like to thank Dr. Ivan Yeung for providing the VX2 tumor model, Dr. Sandy Pang for helpful comments and discussions with regards to pharmacokinetics modeling and the University Health Network animal care staff for their technical services.

Chapter 5. iposome Contrast Agent for CT-based Detection and Localization of Neoplastic and Inflammatory Lesions in Rabbits: Validation with FDG-PET and Histology

89

90

5.1.

Foreword The localization, biodistribution and kinetics of liposomes to healthy and tumor

tissues were quantified using CT in a rabbit model in the previous chapter. Over the course of the study reported in the previous chapter, it was observed that additional lesions (away from the primary tumor site) were enhancing on CT following liposome agent administration. The goal of this chapter is to characterize the CT contrast enhancing lesions (primary or otherwise) with FDG-PET and histology, and assess the performance of liposome-CT to detect these abnormalities with respect to established methods (FDG-PET and histology). The following chapter is currently being considered for publication by Radiology: Zheng J, Allen C, Serra S, Vines D, Charron M, Jaffray DA. Liposome Contrast Agent for CT-based Detection and Localization of Neoplastic and Inflammatory Lesions in Rabbits: Validation with FDG-PET and Histology. Submitted to Radiology on September 1, 2009 (submission number RAD-09-1635, under review).

5.2.

Introduction Non-invasive imaging techniques play an integral role in whole body screening of

neoplastic and inflammatory lesions. Earlier detection, diagnosis and accurate staging will increase the efficacy of disease management [208]. Approaches to implement this include the development of high sensitivity and high resolution imaging devices (i.e. PET, MR and CT), and the engineering of novel agents that enhance the contrast of disease lesions through physiological or biological targeting.

91 Sites of tumor and inflammation are both characterized by enhanced vascular permeability [209]. The unique size and particle surface characteristics of colloidal agents, such as PEGylated liposomes, allow them to be retained in healthy blood vessels, circulate for a prolonged period of time in vivo through avoidance of uptake by the monophagocytic system (MPS), also known as the reticulo-endothelial system (RES), and preferentially extravasate at sites of enhanced vascular permeability. Specifically, it has been reported that liposomes in the 70 to 200 nm size range are optimal for minimizing liver and spleen accumulation, the two main organs that make up the MPS [9]. As a result, liposomes have been explored for both imaging and therapeutic delivery to both sites of tumor and inflammation [210, 211]. Innovations in nanoparticle-based contrast agent development will also have important implications for research and development of novel chemotherapeutic and antiinflammatory drug delivery carriers. For example, nano-sized liposomes have been widely employed as a colloidal carrier for a variety of anti-tumor drugs: liposomal doxorubicin, Doxil® (Johnson & Johnson, Langhorne, PA, USA), being the most clinically successful nanosystem [212]. The ability to non-invasively quantify the concentration of liposomes at tumor sites and to non-invasively map their intratumoral distribution with respect to functional and physiological parameters of the tumor microenvironment have the potential to bring greater insight to the rational development of colloidal therapeutic agents. The recent development of a combined liposome-based CT and MR agent and the characterization of its stability [61], pharmacokinetics and biodistribution [62], as well as imaging performance in both healthy [62] and tumor-bearing animals [63] have been reported. In addition, the tumor accumulation and clearance kinetics of this liposome agent

92 were quantified, which allowed for definition of the optimal tumor imaging window following an intravenous administration [63]. This study assesses the ability of the same liposome agent, together with CT imaging, to localize both tumor and inflammatory sites in a rabbit model with VX2-carcinoma and immune myositis. Finally, the sensitivity of the liposome-CT detection of abnormal lesions is compared to that of FDG-PET. As well, comprehensive histopathological information is reported for tissue samples.

5.3.

Materials and Methods Animal Model of Tumor and Inflammatory Lesions.

All experiments were

performed under an approved animal care and use protocol (University Health Network). Nine healthy male New Zealand White rabbits (Charles River, Wilmington, MA, USA) weighing 2.7 ± 0.3 kg were inoculated with 400 µL of a cell suspension (approximately 107 cells per ml) which was obtained from three VX2-carcinoma-bearing propagation rabbits [213-216]. Specifically, the donor rabbits were sacrificed with a lethal dose (160 mg/kg) of intravenously administered sodium pentobarbital solution (Euthanyl; Bimeda®-MTC Animal Health Inc., Cambridge, Ontario, Canada). Their tumors were excised and placed in Hanks Balanced Salt Solution (HBSS, Sigma-Aldrich, Oakville, Ontario, Canada). In a sterile laminar down-flow air system, the tumor was then washed twice with sterile HBSS and the viable portion of the tumor was cut into small fragments (2 x 3 mm). The final cell suspension was obtained by pushing the tumor fragments through a 70 µm cell strainer in the presence of HBSS. The cell suspension was then loaded into syringes and injected intramuscularly into the animals’ left lateral quadriceps. No attempt was made to isolate the VX2 carcinoma cells from the stromal cells that were part of the donor tumors. This injection

93 resulted in the formation of primary tumors in all nine rabbits. In addition, inflammatory lesions were identified in the skeletal muscles of six of the nine recipient animals and these were classified as non-infectious immune myositis by two human pathologists (S. S. and S. A., Toronto General Hospital, Toronto, Ontario, Canada) with extensive animal histology experience. The mechanism for this focal inflammatory lesion formation is unknown. However, they exclusively occur in tumor-bearing rabbits and their origin may be associated with the exposure to the stromal component of donor tumor. Discussion with a team of veterinary pathologists (P. T. and team, Ontario Veterinary College, University of Guelph, Guelph, Ontario, Canada) confirmed the nature of the inflammatory lesions (i.e. nonsuppurative myositis and necrosis) and a hypothesis pertaining to their pathogenesis was formulated. Specifically, the spontaneous formation of the myositis lesions may be explained via simple host rejection of the tumor with activation of various clonal populations of T lymphocytes as well as dendritic cells and macrophages. The donor tumor cocktail administered to the recipient animals contained immune cells expressing different major histocompatibility complex (MHC) antigens from the donor rabbit. This led to the activation of host (recipient) T cells and dendritic cells, as well as initiation of the nonsuppurative inflammation and eventual necrosis seen around the primary tumors 13 days post-inoculation. In addition, this intense inflammation also caused fragmentation of myofibers and led the activated clonal populations of host T cells as well as macrophages to be directed against epitopes on muscle proteins. Finally, some of these inflammatory cells may “escape” the original site of inflammation in some animals, and set up similar muscledirected inflammation elsewhere in the body, similar to an autoimmune reaction. Regardless

94 of the process, these inflammatory lesions were employed as a model of inflammation in assessing the performance of the liposome agent.

Liposome-CT and FDG-PET Imaging. As in a previously published method [61], 80 nm liposome CT contrast agent co-encapsulating iohexol (Omnipaque®, GE Healthcare, USA) and gadoteridol (Prohance®, Bracco Diagnostic Inc., Princeton, NJ, USA) was prepared and characterized. This liposome agent stably retains the co-encapsulated agents and exhibits a long vascular half-life on the order of 60 hours in VX2-carcinoma bearing rabbits [63]. For this study, the liposome agent was administered intravenously to the rabbits once tumor lesions were established (seven days post inoculation). The rabbits were first induced using a nose cone with a mixture of isoflurane and oxygen via inhalation, and then intubated and maintained under the same inhalant anaesthesia. A slow bolus injection (0.5 mL/second) of 15 mL of the liposomal contrast agent formulation was administered through the marginal ear vein catheter. Each rabbit received 400 ± 80 mg/kg of iohexol (equivalent to 185 ± 37 mg/kg of iodine, corresponding to approximately one third of the iodine dose/body weight typically administered to patients in a bolus form) and 25 ± 4 mg/kg gadoteridol (equivalent to 7 ± 1 mg/kg of gadolinium) co-encapsulated within the liposomes. The PET/CT imaging session (GE Discovery ST PET/CT, General Electric Medical Systems, Milwaukee, WI, USA) took place twelve days after the tumor inoculation procedure, five days post liposome contrast administration and one hour post 18F-FDG injection (30.3 ± 5.1 MBq/kg via the marginal ear vein). The five-day delay was selected based on a previous kinetic study [63] which determined the maximum tumor-to-muscle CT signal ratio was achieved between days five and seven post liposome administration. Per conventional FDGPET imaging protocol, the rabbits were fasted for 18 hours prior to the imaging session, and

95 the blood glucose level for each rabbit was measured immediately before FDG administration using a glucometer (Ascensia® Contour®, Bayer HealthCare LLC, Mishawaka, IN, USA). The mean blood glucose level for the nine rabbits was within the normal range (4.0 ± 1.4 mmol/L). During the one-hour waiting time between the FDG injection and the PET/CT imaging, the rabbits were kept under anaesthesia with the isoflurane and oxygen mixture delivered via inhalation. Two sets of CT images (high and low resolution) were acquired at 80 kVp and 200 mA. One lower resolution CT data set (0.98 x 0.98 x 3.3 mm3 voxel size) was acquired immediately before the PET scan (2D acquisition mode, four bed positions, four minutes of imaging time per bed position and three slice overlap) and was used for attenuation correction of the PET data set (reconstructed at 0.98 x 0.98 x 3.3 mm3 voxels size). In the CT data set used for PET attenuation correction, the tumor/inflammation voxel with the highest iodine contrast accumulation measured 298 HU. According to a published phantom study [217], this level of CT attenuation value would only induce a 7% bias relative to 0 HU in the PET emission data when compared to the

68

Ge-

derived correction. Therefore no adjustment was performed to compensate for the presence of iodine in the PET attenuation correction. The higher resolution CT data set (voxel size of 0.39 x 0.39 x 0.63 mm3) used for image analysis was acquired immediately following the PET acquisition. Figure 5.1 illustrates the experimental timeline in a flow diagram.

Abnormal Lesion Collection and Histopathological Examination. Approximately 18-20 hours post PET/CT imaging, the rabbits were sacrificed with a lethal dose (160 mg/kg) of intravenously administered sodium pentobarbital solution. Based on the CT data sets, needles were inserted into the carcasses to mark the rough anatomical location of the abnormal lesions. The carcasses with needles were then CT scanned and the locations of the

96 lesions with respect to the needle fiducials were measured. Once identified, the lesions were dissected by a certified veterinarian (University Health Network Animal Research Centre) and immediately fixed in approximately 20 times volume excess of 10% formalin. Serial sectioning and staining with hematoxylin-eosin (H&E) and pan-cytokeratin (pan-CK, marker for VX2 carcinoma cells in muscle, as they are of epithelial origin) were performed by the Pathology Research Program at the Toronto General Hospital (Toronto, Ontario, Canada). The tissue slides were scanned at 20X magnification (0.5 µm/pixel) with ScanScope XT (Aperio Technologies Inc., Vista, CA, USA). The histopathological examination was performed by a pathologist (S. S.) and confirmed by a second pathologist (S. A.).

Image Analysis. Lesions were contoured in 3D on the full resolution CT data set using a threshold-based contouring method (Microview v2.2, GE Healthcare). The signal threshold was 3σ greater than the mean HU of a muscle volume in the contra-lateral side. The resulting contours were visually inspected and manually adjusted to exclude bone and assure tumor inclusion. The registration of the lower resolution CT and PET data set was performed using the MIPAV software (CIT, NIH, Bethesda, USA). PET-based assessment of the disease burden was performed by an experienced nuclear medicine physician (M. C.) using the Xeleris visualization station (GE Medical Systems, Milwaukee, WI, USA). The suspect lesions were first identified on the maximum intensity projection of the PET data, and then the triangulation tool was used to verify the lesion on the coronal PET image. To minimize potential bias caused by the presence of the CT agent, the low resolution CT data set was provided and the window and level on the CT data set was preset such that any liposome contrast uptake was imperceptible. Due to the small size of some lesions, partial volume correction was applied to the PET data set [218]. Specifically, the partial volume

97 effect correction was calculated from a separate phantom study (results not shown) where 12 spheres of volumes ranging between 0.18 cm3 and 26.52 cm3 were filled with the same concentration of radioactivity relative to a uniform background (sphere to background ratio of 4:1) and imaged on the same scanner using the same 2D acquisition protocol.

VX2 carcinoma inoculation at the left lateral quadriceps (n=9) 7 days Liposome contrast administration (185 ± 37 mg/kg of iodine) 5 days FDG injection (30.3 ± 5.1 MBq/kg) 1 hour PET/CT imaging 18-20 hours Tissue collection

Figure 5.1 Flow chart illustration of the experimental steps.

98

5.4.

Results Tumor Detection. As summarized in Table 5.1, both liposome-CT and FDG-PET

independently detected the presence of neoplastic lesions, which were histologically confirmed to be malignant, in the lateral upper left thighs of all nine rabbits (sites of tumor inoculation). These primary tumors consisted of medium sized cells with scant cytoplasm and round to oval nuclei arranged in glandular structure and/or in solid sheets. Focal inflammation mainly characterized by lymphocytes in variable proportion was present mostly at the periphery of the lesion. Necrosis was present in all lesions. The VX2 neoplastic cell population within all of the primary tumor lesions were immuno-reactive for pan-cytokeratin. Figure 5.2 illustrates three primary tumors of different sizes. Due to the highly heterogeneous nature of this tumor model, the difference in volume (measured from the CT data set) of the primary lesions ranged from as small as 0.07 cm3 to as large as 7.01 cm3 (i.e. 100-fold difference).

Inflammatory Lesion Detection. In addition to the primary tumors, 25 abnormal lesions were detected in six of the nine rabbits (from the three different tumor inoculations) from the PET/CT imaging study. Thirteen out of the 25 lesions were collected from the animals under CT-guidance and sent for histopathological examination. The remaining lesions were not collected because of their inaccessibility due to their anatomical location. As outlined in Table 5.1, all 7 FDG-PET positive muscle lesions were also seen as abnormal on the CT data set. Six out of these seven lesions were dissected and classified as inflammatory by the pathologists. Figure 5.3 illustrates examples of one inflammatory lesion in the lower lumbar muscle region and another one in the arm. Of the remaining 18 liposome-CT enhanced muscle lesions, 7 were resected and classified as inflammatory by the pathologists.

Volume Range (mm3)

Mean Volume (mm3)

Lesions Detected by FDG-PET (SUVmax Range)

Lesions Detected by Liposome-CT (HUmax Range)

Primary tumors (histology confirmed)

70.1 – 7007.7

3513.7 ± 2763.5

9 (1.5-10.9)

9 (173-596)

Inflammatory lesions (with histology) Inflammatory lesions (no histology)

13.5 – 2732.1 28.8 – 503.5

737.3 ± 814.9 107.6 ± 134.5

6 (2.7-7.1) 1 (3.5)

13 (254-493) 12 (208-498)

Table 5.1 List and classification of the neoplastic and inflammatory lesions detected using CT and PET imaging, their volumes and maximum signal values. Note that all volumes and HUmax values reported here were measured using the full resolution liposome-CT data set (voxel size of 0.39 x 0.39 x 0.63 mm3).

99

100 In general, the inflammatory masses were embedded deep within the muscles of the upper thigh (always superior to the primary tumor and sometimes on the contralateral side), the lower lumbar muscles and the arm muscle. Microscopically, these intramuscular lesions showed irregular margins and consisted of multifocal conspicuous inflammation, infiltrating within the endomysium. The inflammatory infiltrate included numerous histocytes, giant cells, scattered lymphocytes and rare plasma cells. Degenerating muscle fibers and focal calcifications were also present. All of the inflammatory lesions were negative for pancytokeratin, in contrast to the primary tumors which were pan-cytokeratin positive. The pathologists concluded that these inflammatory lesions appear to be non-infectious, and likely a form of immune myositis.

Comparison of Liposome-CT Accumulation and FDG-PET Uptake. In this pilot study, liposome-CT and FDG-PET were both able to detect all of the nine primary tumor lesions. The difference in the performance of the two imaging techniques to identify inflammatory lesions was not necessarily a function of tumor size. Specifically, liposome-CT had a higher sensitivity for detecting muscle inflammations (25 lesions vs. 7 lesions detected by FDG-PET), with volumes ranging between 0.01 cm3 and 2.73 cm3. The lesions that were also detected by FDG-PET ranged in volume between 0.05 cm3 and 1.04 cm3. While the lower spatial resolution may be the major contributor for PET’s inability to detect the eight lesions that were less than 0.05 cm3 in volume; the remaining ten lesions that were not detected by PET ranged between 0.06 cm3 and 2.73 cm3. For the latter cases, their proximity to high FDG uptake structures combined with an overall lower FDG uptake may be contributing factors for missed abnormality identification. An attempt was made to determine whether it is possible to discriminate between the two types of abnormalities (tumor and

101 inflammation) based on imaging signal alone. Figure 5.4a illustrates that five days postadministration, the degree of liposome accumulation may allow for potential differentiation of neoplastic from inflammatory lesions. In general, VX2 tumors exhibit less CT contrast enhancement compared to inflammatory lesions. The mean of the ratios of HUmean measured in the tumor over the HUmean measured in the blood (aorta) obtained from 14 distinct VX2tumor bearing rabbits (nine rabbits from this study + five additional rabbits from a separate study) was 0.85 ± 0.11. Conversely, the mean of the ratios of HUmean measured in the inflammatory sites (25 lesions from the 9 rabbits from this study + 8 lesions from the 5 additional rabbits) over the HUmean measured in the blood obtained from the rabbits employed in this study was 1.31 ± 0.25. Results from Welch’s t test confirmed that the HUmean value from the CT data set obtained 5-days post-liposome administration can be used to differentiate between neoplastic and inflammatory lesions (p < 0.0001). Although the average size of tumor lesions (5.97 ± 4.52 cm3) was much greater than the average size of the inflammation lesions (0.40 ± 0.60 cm3), there is no clear correlation between liposome uptake and lesion size (Figure 5.4b).

Figure 5.4c depicts the range of partial volume

adjusted SUVmax values for the two lesion types calculated from the FDG-PET data set. The mean adjusted SUVmax for the nine tumor lesions and eleven inflammatory lesions were 4.9 ± 2.0 and 5.3 ± 2.3, respectively. Welch’s t test concluded that there is no significant difference in the adjusted SUVmax values obtained from these two lesion types (p > 0.15). A CT study on a separate group of four rabbits (different from the nine employed for this CT/PET investigation) was conducted to investigate the difference in liposome accumulation and clearance kinetics from tumors and inflammatory lesions. Figure 5.5 reports the results from the kinetic study and it illustrates that not only the inflammatory lesions exhibit higher CT

102 contrast enhancement over time, but also the greatest difference between the mean CT signal measured in tumor and inflammatory sites occurred at 5 days post liposome administration. It is also interesting to note that there was insufficient contrast uptake at the sites of inflammation in the first 48 hours post injection.

Incidental Findings. A total of two lymph nodes (0.10 cm3 and 0.12 cm3 in volume) were classified as malignant from the FDG-PET results (Figure 5.6a and 5.6b). However, neither was contrast enhanced in the CT data set. One of these two lymph nodes was successfully collected and was determined to be unequivocal nodal metastasis (pancytokeratin positive) by the pathologists. In a separate case (not from this series of animals, Figure 5.6c), a VX2 carcinoma-bearing rabbit was administered the same liposome agent and imaged with CT for five days post-injection and sacrificed. An enlarged para-aortic lymph node (long axis measures 1.5 cm) at the lower lumbar muscle level and of abnormal morphology was collected, confirmed to be malignant and in this case was also not contrastenhanced in CT. Although publications have reported liposome accumulation in malignant lymph nodes following interstitial administrations [124], the above findings suggest that this liposome formulation does not significantly accumulate in lymph nodes that have been invaded by tumor cells following intravenous administration. However, due to the low numbers of animals, no definitive conclusion can be made. A suspect liver lesion (no histology available) of 1.15 cm3 in volume was also detected by FDG-PET (SUVmax = 4.6). The same lesion was shown as a hypo-intense region in the liposome-CT data set. This is consistent with the imaging morphology of hypovascular liver tumors detected with a contrast-enhanced CT protocol [219]. Lastly, three suspect bone lesions were identified in the FDG-PET data set with SUVmax values ranging between 1.7 and 2.2. No histology samples

103 are available to confirm the nature of these three lesions, and the liposome-CT data set does not show an elevated contrast uptake pattern in these regions.

Figure 5.2 Three cases of primary tumors detected by CT and PET, and confirmed by histology (pan-CK positive). The volumes of the lesions depicted are 7007 mm3, 299 mm3 and 70 mm3 for cases A, B and C, respectively.

104

Figure 5.3 Two cases of inflammatory lesions in the muscle detected by CT and PET, and confirmed by histology (pan-CK negative). The volumes of the lesions depicted are 2732 mm3 and 173 mm3 for cases A and B, respectively. Lesion A was located in the right lower lumbar muscle and lesion B was located in the left arm.

105 (a) 2.2

Lesion to blood HUmean ratio

2.0 1.8 1.6 75%

1.4 50%

1.2 25%

1.0 75% 50% 25%

0.8 0.6 0.4

Tumor

Inflammation

(b) 2.25

Inflammation Tumor

Lesion to blood HUmean ratio

2.00

1.75

1.50

1.25

1.00

0.75

0.50 0

2

4

6

8

10

12 3

Lesion size (cm )

14

16

18

20

106 (c) 12

Adjusted SUVmax

10

8

75% 75%

50%

6 50%

25%

4 25%

2

0

Tumor

Inflammation

Figure 5.4 Imaging signal intensities of neoplastic and inflammatory lesions: a) ratio of mean HU measured in an entire lesion over that measured in the blood from the liposome-CT data set; b) difference in the normalized CT signals between the tumor and inflammatory sites is independent of lesion size; and c) partial volume effect adjusted SUVmax measured in the FDG-PET data set for the two lesion types. The upper, middle and lower bounds of the boxes indicate the 75th, 50th and 25th percentile, and the dash inside each box marks the mean value of each sample population. Note that 5 additional VX2-carcinoma bearing rabbits with 8 inflammatory lesions from a separate study were included in the CT data set (a).

107

200 175

Lesion HUmean

150 125 100 75 50 25

Tumor (n=4) Inflammation (n=8)

0 0

50

100

150

200

250

300

350

Time (h)

Figure 5.5 Kinetic profiles of liposome contrast agent accumulation and clearance in tumor and inflammatory lesions. The measurements were performed on four male New Zealand White rabbits each bearing a primary VX2 carcinoma tumor in their upper left thigh and each exhibiting between one to three inflammatory lesions in the lower lumbar muscle area. There are no data points before 48 hours post liposome injection for the inflammatory lesions because there was insufficient contrast enhancement to identify the lesion from surrounding muscle.

(a)

(b)

(c)

Figure 5.6 a) and b) Two examples of FDG-PET positive lymph nodes (5 mm and 6 mm) that were not detected using liposome-CT; c) a malignant lymph node excised from an animal from a separate study. The node (15 mm) was not contrast-enhanced in CT.

108

109

5.5.

Discussion The goal of this investigation was to demonstrate the ability of a newly developed

liposomal CT agent to localize in both tumor and inflammatory sites via the well-described enhanced permeation and retention (EPR) effect [18]. Although liposomes have been extensively reported to deliver drugs and radionuclide imaging agents to these sites [220, 221], this is the first time that CT-based assessment of liposome accumulation in both neoplastic and inflammatory lesions is shown. Furthermore, no studies have sought to compare the performance of an EPR targeting nanoscale contrast agent with that of FDGPET. Image-based differentiation of malignant tumor and inflammatory lesions has proven to be challenging due to physiological similarities present in these two types of abnormalities. For example, both tumor and inflammation are generally characterized by increased vascular permeability, metabolic activity and presence of immune cells such as macrophages. As a result, in this study, it was not unexpected that both types of lesions were detected by liposome-CT and FDG-PET. However, it is important to note that liposomes and FDG target abnormal lesions through two very distinct bio-physiological processes. The liposomes employed in this study were spherical particles of roughly 80 nm in diameter [61]. Due to their size, they are retained within the endothelial walls of normal vasculature, but are able to leak into the interstitial space of tissues through the highly permeable vessels that characterize both VX2 tumors and inflammatory sites. Furthermore, the slow venous return and poor lymphatic drainage system found at tumor sites, as well as uptake by tumor and inflammation associated macrophages significantly slows down their clearance [15, 23, 222]. The combination of these factors results in a preferential accumulation and retention of

110 liposomes at tumor sites. On the other hand, FDG has a molecular weight of 181.3 Daltons. It is approximately 104 times smaller than a single CT liposome employed in this particular study. Because its size matches that of a small molecule agent, it is subject to fast renal clearance and does not localize at sites of tumor and inflammation via the EPR effect. However, its retention within cells of high metabolic activity (i.e. neoplastic cells with high proliferation rate and activated macrophages) allows it to be used effectively for tumor and inflammation imaging [223-225]. Reports from a number of research groups have demonstrated that dynamic imaging of liposomes [226] and FDG [225, 227-229] at tumor and inflammatory sites allows for successful classification of the two lesion types. In this investigation, only one imaging time point was used to differentiate tumor from inflammation. For the particular tumor and inflammation model employed in this study, liposomes demonstrated a statistically significant difference in their mean accumulation relative to blood at the two sites, while FDG did not. It is interesting to note that studies that have investigated the microdistribution of liposomes and FDG in neoplastic lesions have concluded that both agents also localized in the non-tumor component of the tumor tissue. Specifically, Kubota et al. reported in their micro-autoradiography study that the macrophage layer surrounding tumor necrosis, the young granular tissue with capillary vessels, fibroblasts and macrophages surrounding the tumor mass, as well as the necrotic area with macrophages all showed higher grain count than in the tumor cells themselves [230]. However, in human and animal tumors, macrophages usually constitute 20-30% of the cellular tumor mass, therefore it is reasonable to conclude that most of the radioactivity of 18F-FDG in tumors originates from viable tumor cells [229, 230]. Therefore, it is not surprising that inflamed tissues also exhibit high FDG

111 uptake. Kaim et al. reported in their autoradiographic quantification of

18

F-FDG uptake in

experimental soft-tissue abscesses in rats that the FDG uptake clearly coincided with the presence of macrophages, and that no substantial increase in FDG uptake was detected within the fibroblast-enriched granulation tissue [231]. Conversely, a study investigating the microdistribution of liposomes within C-26 colon carcinoma tumors reported that although the gold-labeled liposomes were seen to be predominantly scattered in the extracellular space between tumor cells, they also localized in areas surrounding blood vessels, as well as in the endosomes and secondary lysosomes of tumor-associated macrophages [232]. Similarly, in a rat model of focal infection, Laverman et al. found liposomes in the vicinity of blood vessels and some present in endothelial cells, as well as significant localization within macrophages [222]. Therefore, it could be hypothesized that the difference measured in this study between the mean HU of tumor and inflammatory lesions was a function of both microvessel density and macrophage presence. The difference in the liposome uptake kinetics at the sites of tumor and inflammation (Figure 5.5) seem to indicate that it is the interaction between liposome and macrophages that play a greater role in modulating its accumulation in inflammatory lesions. The tumor kinetics curve peaks at 48 hours (two days) post liposome administration (indicating that EPR occurs during this earlier time window), while the inflammation kinetics curve peaks at 120 hours (five days) post injection (indicating that a process different from EPR dominates this later time window). Current efforts are in place to quantify the microvessel and macrophage density from the histology obtained for each lesion in order to confirm the hypothesis. In summary, results from this investigation demonstrated that increased contrast of neoplastic and inflammatory lesions in CT is achieved with the administration of a liposomal

112 nano-agent. Its differential accumulation at sites of tumor and inflammation, as well as its higher sensitivity in the detection of soft-tissue inflammatory lesions compared to FDG-PET, suggest that, if approved for human use, liposomes could play a role in CT-based disease screening and image-guided biopsy.

5.6.

Acknowledgements This work is funded in-part by CIHR and OICR research grants and the Fidani Chair

in Radiation Physics. Jinzi Zheng is grateful for the CIHR Banting and Best Canada Graduate Scholarship. The authors would like to thank Dr. Anguo Zhong and Dr. Margarete Akens for performing the VX2 tumor propagation, Dr. Alyssa Goldstein and Sandra Lafrance for assistance during tissue collection, Dr. Harald Keller for providing the PET phantom data used for partial volume correction, and Dr. Sylvia Asa and Dr. Patricia Turner for their contribution to the histopathology component.

Chapter 6. Summary and Future Directions

113

114

6.1.

Summary A nano-sized liposome formulation was developed over the course of this thesis with

the ability to stably carry multiple imaging moieties. Extensive in vitro and in vivo characterization was performed to explore the feasibility and utility of employing this nanosystem for a number of applications including multimodality CT and MR imaging (chapter 2), longitudinal vascular imaging (chapter 3), CT-based assessment of liposome biodistribution and kinetics (chapter 4), and CT detection of tumor and inflammatory lesions (chapter 5). The three main innovations of this thesis are: 1) development of the first colloidal dual CT and MR imaging agent; 2) employment of CT as a quantitative and noninvasive method for longitudinal evaluation of the biodistribution and kinetics of a nanoparticulate agent; and 3) demonstration that the use of the liposome agent in conjunction with CT imaging has superior performance, compared to FDG-PET, in the detection of abnormalities of inflammatory origin as well as differentiation of these lesions from those of neoplastic origin. The work conducted within this thesis is highly inter-disciplinary. It took advantage of the arguably most well characterized pharmaceutical carrier, the liposome, optimized a formulation that can stably co-encapsulate multiple hydrophilic contrast agents and explored its use in different imaging applications. The liposome literature goes back to the 1960s, when it was first used as a model for studying biological membranes [233], then it was recognized to be a promising carrier for pharmaceutical drugs and imaging agents. In medical imaging, extensive investigations were first performed in nuclear medicine to assess its suitability both as a radiopharmaceutical delivery system (i.e. by Caride et al. in 1976 [234]) and as an imaging agent [235]. The versatility of liposomes allowed for extension of their

115 employment in other imaging modalities such as CT (first preclinical report by Havron et al. in 1981 [236] and first human study published by Leander et al. in 1998 [112]), MR (Magin and et. reported the use of paramagnetic liposomes in mice in 1986 [237]), ultrasound (Unger et al. described the feasibility of nitrogen-filled liposomes for vascular imaging in 1992 [238]) and optical imaging. The concept of using liposomes to co-load multiple imaging moieties for cross-modality imaging emerged at around 2005 and 2006 with publications reporting successful engineering of systems suitable for dual MR and optical imaging (Mulder et al. in December 2005 [239]), MR and CT imaging (Zheng et al. in March 2006 [61]), and MR and SPECT imaging (Zielhuis et al in December 2006 [240]). Although Chapter 2 of this thesis constituted one of the pioneer reports of multimodality imaging using liposomes, other research groups have investigated other nano-systems, such as nanoparticles [105, 241-243], nanocrystals [244], lipo-proteins [245] and dendrimers [246] for the same purpose. It is likely that while a wide range of different nanosystems will remain to be used in preclinical research applications, the first nanosystem that will achieve clinical approval for use in humans will be explored for employment across different applications that benefit from the use of multimodality imaging.

6.2.

Future Directions There are three broad directions in which further investigation can be pursued: 1)

comprehensive toxicity and efficacy studies to be conducted in appropriate animal models in compliance with the regulatory approval application process in parallel with the selection of a suitable commercialization partner; 2) continued innovation in the agent development side to add functionality to the existing liposome platform (i.e. imaging capability beyond CT and

116 MR, specific cell targeting, ability to respond to triggers); and 3) employment of this nano-system to further increase understanding of liposome transport, distribution and disease localization, especially how these are modulated by the patho-physiology of the biological system under investigation.

6.2.1. Technology

Translation

and

Commercialization:

Challenges

and

Opportunities The CT and MR liposome contrast agent developed within the framework of this thesis demonstrated high stability in vitro and in vivo, no significant toxicity, and it showed potential to benefit a number of imaging applications. Additional studies (results not reported in this thesis) investigated its storage shelf-life and found that the liposomes in a physiological buffer solution can be stored for at least one month at both room temperature and 4°C without statistically significant leakage of the encapsulated agents or change in the mean liposome size and size distribution. Furthermore, it was feasible to scale-up the liposome contrast agent production from 20 mL per batch to 200 mL per batch through the use of a large-scale LipexTM high-pressure extruder (Northern Lipids Inc., Burnaby, BC, Canada) without increasing the production time and at the same time maintaining the same physico-chemical characteristics of the liposome solution reported in this thesis. In order to further the commercialization and technology translation efforts, the current liposome contrast agent production procedure must be adapted to comply with the Good Manufacturing Practice (GMP) standards. As final stage sterilization techniques, such as irradiation, heat, high pressure and filtration are not suitable for liposomes, a

117 GMP production facility must be constructed to include a clean room where the entire liposome preparation process will take place using sterile materials and production instruments. Initial discussion with the Canadian regulatory agency, Health Canada, has indicated that further toxicity, clearance and metabolism studies should be conducted in two animal models (one rodent and one non) in order to obtain comprehensive information on the safety profile of this new agent. Furthermore, at least one clinical application needs to be identified, with the support from clinicians, in order to allow for patient population and clinical endpoint selection during the efficacy assessment stage of the clinical trials. It is challenging for an academic group to gather resources and expertise for the translation and commercialization of a medical technology. Joint efforts with the UHN business development office have led to successful funding applications through both governmental and non-governmental sources to support these activities through task outsourcing to pharmaceutical safety evaluation companies and regulatory approval consultants.

If regulatory approval for use of this liposome agent is achieved, this

research thesis will become a true example of successful translation from bench to bedside.

6.2.2. Extension to a Modular Multimodality Imaging Platform In routine clinical practice, specific imaging modalities are employed at different stages of disease diagnosis and treatment. For example, combinations of anatomical and functional imaging techniques such as CT, MR, PET, SPECT and US are often utilized

118 for disease diagnosis, staging and treatment planning. X-ray, MR, US and optical imaging systems are also found in the treatment room to guide the delivery of various interventions such as surgery, radiotherapy and radiofrequency ablation with the aim of improving the accuracy of these procedures [78, 79, 247-251]. The benefit of combining the different imaging modality is that they can provide complementary information on the anatomy, physiology and function of the biological system under investigation. However, because they rely on different signal generating mechanisms, it is sometimes difficult to identify common structures for spatial reference. Accurate definition of the anatomical location and boundary of the biological target is essential for therapy planning and delivery. As a result, it would be extremely advantageous if signal generating agents having the same pharmacokinetics, biodistribution and disease targeting ability are available and their co-localization across imaging modalities would increase the confidence in identification of disease and its boundary. A viable strategy is to engineer a nanoplatform-based approach, which ensures that the biological performance of the nanoagent remains constant while slight inert modifications can be made to custom match the signal generating moiety load to the imaging modality or modalities of choice. Liposomes are highly versatile colloidal particles that allow for encapsulation of hydrophilic molecules within their aqueous core, the incorporation of hydrophobic molecules within their lipid bilayer and insertion of additional lipid groups within their bilayer that can either conjugate molecules or metal chelators [220]. In parallel with the development of the dual-molecule co-encapsulation strategy described in this thesis, our research team also explored the lipid exchange technique. Specifically, the incorporation of lipid groups into the pre-formed liposome bilayer was optimized. These lipid groups

119 are either directly conjugated to an optical dye (i.e. Cy 5.5) or attached to a metal chelator that binds to either a positron or single photon emitter (i.e.

64

Cu or

111

In), as well as

incorporate an active targeting ligand on the distal end of the PEG group to modulate the distribution of liposomes to specific cell populations. Figure 6.1 shows the schematic representation of the modular liposome multimodality imaging platform currently under development. The resulting platform is aimed at providing persistent and co-localized signal enhancements across multiple imaging systems (CT, MR, PET, SPECT and optical). It has the potential to seamlessly bridge wide ranges of spatial, temporal and sensitivity scales and to be employed throughout a variety of clinical scenarios (i.e. diagnostic, pre-operative, intra-operative and follow-up imaging). Furthermore, strategies to activate or deactivate different components of the liposome system following either endogenous biological queues (i.e. temperature, pH, matrix metalloproteinase levels) or external triggers (i.e. irradiation, hyperthermia, magnetic field) can be explored to engineer a “smart” modular multimodality imaging platform.

Figure 6.1 Schematic representation of the modular multimodality liposome imaging platform. The aqueous core of the liposome can be used to cargo hydrophilic small molecule therapeutic and imaging agents. Physical attachment of polymers such as DTPA, DOTA and HYNIC onto phosphatidylethanolamine (PE) lipids allows for chelation of PET and SPECT metal radioisotopes. Optical imaging probes and active targeting ligands can be either directly conjugated onto the PE lipid or the distal end of the PEG chain, or be incorporated into the liposome system through strepavidin and biotin binding. 120

121

6.2.3. Additional Characterization of Liposome Transport and Distribution The successful development of a modular multimodality imaging platform will allow for comprehensive probing of the complexity of tumors. For example, Chapters 4 and 5 of the present thesis demonstrated that CT is advantageous for high-resolution disease detection and characterization. The voxel size of the clinical CT scanner (Discovery ST, GE Healthcare, USA) employed throughout the rabbit imaging studies reported in this thesis was 390 x 390 x 630 µm3. Recent developments in the area of preclinical imaging resulted in commercially available microCT systems with the ability to perform in vivo imaging with an isotropic voxel size of approximately 15 µm in mice (Inveon microCT, Siemens Preclinical Solutions, USA). As a result, it is now feasible to image the micro-distribution of imaging agents, such as the liposome agent developed here, in tumors in vivo longitudinally. To date,

ex vivo imaging techniques such as autoradiography and confocal/multiphoton microscopy have been routinely employed to quantify the distribution of radio- and optically-labelled agents in tissue sections at tens of microns and sub-micron resolutions, respectively [252, 253] . The benefits of volumetric in vivo assessment of agent micro-distributions are clear. Not only does it allow for mapping of the distribution of the agent of interest in a live biological system without the deformations that often occur during the preparation of tissue sections [254], but it also enables multiple measurements to be performed on the same animal over time, permitting the evaluation of kinetic changes in the micro-distribution pattern. When performing serial high-resolution microCT imaging, the amount of radiation dose delivered to an animal, especially small size rodents such as mice, should be considered. Publications that investigated the effect of serial microCT radiation dose to animals have

122 found that there was no effect on tumor growth, mouse survival or ability to form metastasis [255, 256]. Daibes Figueroa et al. reported in their thermo luminescent dosimeter (TLD) assessment of mouse dosimetry an absorbed dose of 76 ± 5 mGy for one full rotation (360°) of microCT scan (MicroCAT IITM, Siemens Preclinical Solutions, USA) at 80 kVp and 54 mA with 0.5 mm of aluminum filtration [257]. This is equivalent to approximately 1/5 of the whole body absorbed dose delivered to a 20 g mouse with the administration of one of the longer lived SPECT isotope

111

In (360 mGy, assuming no biological washout [258]). These

dose considerations makes it feasible to design serial high-resolution microCT imaging studies. An attractive application of high-resolution imaging is for comparison of the microdistribution and kinetics of passively and actively targeted agents in tumors. To date, many investigations have been carried out to assess the value of ligand-directed nanoparticle delivery to tumors. Increased therapeutic efficacy has been often associated with the employment of an actively targeted agent. However, no consensus has been reached with respect to the underlying mechanisms that resulted in the improved therapeutic ratio. Several groups suggest that the presence of an actively targeted ligand on the surface of nanoparticles does not alter its tumor localization, but rather increases the intracellular uptake of the nanocomplex [259, 260]. Other reports indicate that the actively targeted nanoparticles do have an enhanced accumulation at tumor sites [261-265]. The multi-factorial nature of this comparative investigation makes it difficult to draw straight-forward conclusions [266]. Longitudinal quantitative high-resolution microCT monitoring of the spatial and temporal distribution of targeted and non-targeted nanoparticles within murine tumors at isotropic resolutions of 15 µm is an attractive method that can potentially help elucidate the

123 mechanisms of tumor accumulation and distribution of different actively targeted nanosystems. Furthermore, if the nanosystem is co-labelled with a CT agent and a near infrared optical dye, and it is imaged with a microCT and an in situ confocal imaging probe (i.e. Leica FCM1000), it is possible to assess both the intratumoral localization and the tumor cell uptake of this nanosystem. The study conducted in Chapter 5 illustrated that although both liposomes and FDG significantly accumulate and are retained in neoplastic lesions, they do so according to very different underlying biological and physiological processes. Much research has focused on using imaging agents, also known as imaging biomarkers, to spatially map different environments within a tumor (i.e. viable, necrotic and hypoxic areas [267]), it would be interesting to query the effect of the diverse tumor micro-environment on the distribution and uptake of imaging and therapeutic agents, in particular macromolecular agents such as liposomes. The well-known EPR effect describes the differential accumulation and retention of macromolecules at healthy and tumor tissues resulted from the difference in physiology (i.e. increased vessel permeability and decreased lymphatic drainage found at tumor sites). It is also generally accepted that parameters that define the tumor micro-environment, such as regional micro-vessel density, blood flow, degree of oxygenation, pH and interstitial fluid pressure, can influence the distribution of both imaging and therapeutic agents [268]. In fact, functional imaging techniques such as functional CT and DCE-MR rely on these local changes to derive information in order to classify the different tumor voxels. The employment of macromolecular agents, especially groups of agents of different but welldefined molecular weight and size, with high-resolution imaging modalities such as CT and MR, has the potential to further elucidate the interplay of the different tumor micro-

124 environmental parameters described above. Furthermore, when used in conjunction with other molecular imaging methods (i.e. FAZA-PET for hypoxia imaging), biological mapping of abnormal lesions and high-resolution classification of the tumor micro-environment can be achieved. This will have important implications for cancer staging, treatment planning, delivery guidance, as well as therapy follow-up.

125

References 1.

Li, S.D. and L. Huang, Pharmacokinetics and biodistribution of nanoparticles. Mol Pharm, 2008. 5(4): p. 496-504.

2.

Davis, M.E., Z.G. Chen, and D.M. Shin, Nanoparticle therapeutics: an emerging

treatment modality for cancer. Nat Rev Drug Discov, 2008. 7(9): p. 771-82. 3.

Choi, H.S., W. Liu, P. Misra, E. Tanaka, J.P. Zimmer, B. Itty Ipe, M.G. Bawendi, and J.V. Frangioni, Renal clearance of quantum dots. Nat Biotechnol, 2007. 25(10): p. 116570.

4.

Venturoli, D. and B. Rippe, Ficoll and dextran vs. globular proteins as probes for testing

glomerular permselectivity: effects of molecular size, shape, charge, and deformability. Am J Physiol Renal Physiol, 2005. 288(4): p. F605-13. 5.

Clough, G., Relationship between microvascular permeability and ultrastructure. Prog Biophys Mol Biol, 1991. 55(1): p. 47-69.

6.

Oku, N. and Y. Namba, Long-circulating liposomes. Crit Rev Ther Drug Carrier Syst, 1994. 11(4): p. 231-70.

7.

Gabizon, A. and F. Martin, Polyethylene glycol-coated (pegylated) liposomal

doxorubicin. Rationale for use in solid tumours. Drugs, 1997. 54 Suppl 4: p. 15-21. 8.

Sadzuka, Y., S. Hirotsu, and S. Hirota, Effect of liposomalization on the antitumor

activity, side-effects and tissue distribution of CPT-11. Cancer Lett, 1998. 127(1-2): p. 99-106. 9.

Liu, D., A. Mori, and L. Huang, Role of liposome size and RES blockade in controlling

biodistribution and tumor uptake of GM1-containing liposomes. Biochim Biophys Acta, 1992. 1104(1): p. 95-101. 10.

Romero, E.L., M.J. Morilla, J. Regts, G.A. Koning, and G.L. Scherphof, On the

mechanism of hepatic transendothelial passage of large liposomes. FEBS Lett, 1999.

448(1): p. 193-6. 11.

Levchenko, T.S., R. Rammohan, A.N. Lukyanov, K.R. Whiteman, and V.P. Torchilin,

Liposome clearance in mice: the effect of a separate and combined presence of surface charge and polymer coating. Int J Pharm, 2002. 240(1-2): p. 95-102.

126 12.

Geng, Y., P. Dalhaimer, S. Cai, R. Tsai, M. Tewari, T. Minko, and D.E. Discher, Shape

effects of filaments versus spherical particles in flow and drug delivery. Nat Nanotechnol, 2007. 2(4): p. 249-55. 13.

Matsumura, Y. and H. Maeda, A new concept for macromolecular therapeutics in cancer

chemotherapy: mechanism of tumoritropic accumulation of proteins and the antitumor agent smancs. Cancer Res, 1986. 46(12 Pt 1): p. 6387-92. 14.

Noguchi, Y., J. Wu, R. Duncan, J. Strohalm, K. Ulbrich, T. Akaike, and H. Maeda, Early

phase tumor accumulation of macromolecules: a great difference in clearance rate between tumor and normal tissues. Jpn J Cancer Res, 1998. 89(3): p. 307-14. 15.

Maeda, H. and Y. Matsumura, Tumoritropic and lymphotropic principles of

macromolecular drugs. Crit Rev Ther Drug Carrier Syst, 1989. 6(3): p. 193-210. 16.

Seymour, L.W., K. Ulbrich, P.S. Steyger, M. Brereton, V. Subr, J. Strohalm, and R. Duncan, Tumour tropism and anti-cancer efficacy of polymer-based doxorubicin

prodrugs in the treatment of subcutaneous murine B16F10 melanoma. Br J Cancer, 1994.

70(4): p. 636-41. 17.

Takakura, Y. and M. Hashida, Macromolecular carrier systems for targeted drug

delivery: pharmacokinetic considerations on biodistribution. Pharm Res, 1996. 13(6): p. 820-31. 18.

Maeda, H., J. Wu, T. Sawa, Y. Matsumura, and K. Hori, Tumor vascular permeability

and the EPR effect in macromolecular therapeutics: a review. J Control Release, 2000.

65(1-2): p. 271-84. 19.

Yuan, F., M. Dellian, D. Fukumura, M. Leunig, D.A. Berk, V.P. Torchilin, and R.K. Jain,

Vascular permeability in a human tumor xenograft: molecular size dependence and cutoff size. Cancer Res, 1995. 55(17): p. 3752-6. 20.

Yuan, F., M. Leunig, S.K. Huang, D.A. Berk, D. Papahadjopoulos, and R.K. Jain,

Microvascular permeability and interstitial penetration of sterically stabilized (stealth) liposomes in a human tumor xenograft. Cancer Res, 1994. 54(13): p. 3352-6. 21.

Iyer, A.K., G. Khaled, J. Fang, and H. Maeda, Exploiting the enhanced permeability and

retention effect for tumor targeting. Drug Discov Today, 2006. 11(17-18): p. 812-8. 22.

Skinner, S.A., P.J. Tutton, and P.E. O'Brien, Microvascular architecture of experimental

colon tumors in the rat. Cancer Res, 1990. 50(8): p. 2411-7.

127 23.

Seymour, L.W., Y. Miyamoto, H. Maeda, M. Brereton, J. Strohalm, K. Ulbrich, and R. Duncan, Influence of molecular weight on passive tumour accumulation of a soluble

macromolecular drug carrier. Eur J Cancer, 1995. 31A(5): p. 766-70. 24.

Byrne, J.D., T. Betancourt, and L. Brannon-Peppas, Active targeting schemes for

nanoparticle systems in cancer therapeutics. Adv Drug Deliv Rev, 2008. 60(15): p. 161526. 25.

Debbage, P., Targeted drugs and nanomedicine: present and future. Curr Pharm Des, 2009. 15(2): p. 153-72.

26.

Sawant, R.M., J.P. Hurley, S. Salmaso, A. Kale, E. Tolcheva, T.S. Levchenko, and V.P. Torchilin,

"SMART"

drug

delivery

systems:

double-targeted

pH-responsive

pharmaceutical nanocarriers. Bioconjug Chem, 2006. 17(4): p. 943-9. 27.

Scherer, R.L., M.N. VanSaun, J.O. McIntyre, and L.M. Matrisian, Optical imaging of

matrix metalloproteinase-7 activity in vivo using a proteolytic nanobeacon. Mol Imaging, 2008. 7(3): p. 118-31. 28.

Kong, G. and M.W. Dewhirst, Hyperthermia and liposomes. Int J Hyperthermia, 1999.

15(5): p. 345-70. 29.

Bikram, M. and J.L. West, Thermo-responsive systems for controlled drug delivery. Expert Opin Drug Deliv, 2008. 5(10): p. 1077-91.

30.

Paasonen, L., T. Laaksonen, C. Johans, M. Yliperttula, K. Kontturi, and A. Urtti, Gold

nanoparticles enable selective light-induced contents release from liposomes. J Control Release, 2007. 122(1): p. 86-93. 31.

Xie, W., L. Wang, Y. Zhang, L. Su, A. Shen, J. Tan, and J. Hu, Nuclear Targeted

Nanoprobe for Single Living Cell Detection by Surface-Enhanced Raman Scattering. Bioconjug Chem, 2009. 32.

D'Souza, G.G., S.M. Cheng, S.V. Boddapati, R.W. Horobin, and V. Weissig,

Nanocarrier-assisted sub-cellular targeting to the site of mitochondria improves the proapoptotic activity of paclitaxel. J Drug Target, 2008. 16(7): p. 578-85. 33.

Breunig, M., S. Bauer, and A. Goepferich, Polymers and nanoparticles: intelligent tools

for intracellular targeting? Eur J Pharm Biopharm, 2008. 68(1): p. 112-28. 34.

Working, P.K. and A.D. Dayan, Pharmacological-toxicological expert report. CAELYX.

(Stealth liposomal doxorubicin HCl). Hum Exp Toxicol, 1996. 15(9): p. 751-85.

128 35.

Siegal, T., A. Horowitz, and A. Gabizon, Doxorubicin encapsulated in sterically

stabilized liposomes for the treatment of a brain tumor model: biodistribution and therapeutic efficacy. J Neurosurg, 1995. 83(6): p. 1029-37. 36.

Gabizon, A.A., Y. Barenholz, and M. Bialer, Prolongation of the circulation time of

doxorubicin encapsulated in liposomes containing a polyethylene glycol-derivatized phospholipid: pharmacokinetic studies in rodents and dogs. Pharm Res, 1993. 10(5): p. 703-8. 37.

Amantea, M., M.S. Newman, T.M. Sullivan, A. Forrest, and P.K. Working, Relationship

of dose intensity to the induction of palmar-plantar erythrodysesthia by pegylated liposomal doxorubicin in dogs. Hum Exp Toxicol, 1999. 18(1): p. 17-26. 38.

Colbern, G.T., A.J. Hiller, R.S. Musterer, P.K. Working, and I.C. Henderson, Antitumor

activity of Herceptin in combination with STEALTH liposomal cisplatin or nonliposomal cisplatin in a HER2 positive human breast cancer model. J Inorg Biochem, 1999. 77(12): p. 117-20. 39.

Working, P.K., M.S. Newman, T. Sullivan, and J. Yarrington, Reduction of the

cardiotoxicity of doxorubicin in rabbits and dogs by encapsulation in long-circulating, pegylated liposomes. J Pharmacol Exp Ther, 1999. 289(2): p. 1128-33. 40.

Willmann, J.K., N. van Bruggen, L.M. Dinkelborg, and S.S. Gambhir, Molecular

imaging in drug development. Nat Rev Drug Discov, 2008. 7(7): p. 591-607. 41.

Abbey, C.K., A.D. Borowsky, E.T. McGoldrick, J.P. Gregg, J.E. Maglione, R.D. Cardiff, and S.R. Cherry, In vivo positron-emission tomography imaging of progression and

transformation in a mouse model of mammary neoplasia. Proc Natl Acad Sci U S A, 2004. 101(31): p. 11438-43. 42.

Bhatnagar, A., R. Hustinx, and A. Alavi, Nuclear imaging methods for non-invasive drug

monitoring. Adv Drug Deliv Rev, 2000. 41(1): p. 41-54. 43.

Fischman, A.J., N.M. Alpert, and R.H. Rubin, Pharmacokinetic imaging: a noninvasive

method for determining drug distribution and action. Clin Pharmacokinet, 2002. 41(8): p. 581-602. 44.

Rottey, S., A. Signore, and C. Van de Wiele, Radiolabelled chemotherapeutics. Q J Nucl Med Mol Imaging, 2007. 51(2): p. 139-51.

129 45.

Hamoudeh, M., M.A. Kamleh, R. Diab, and H. Fessi, Radionuclides delivery systems for

nuclear imaging and radiotherapy of cancer. Adv Drug Deliv Rev, 2008. 60(12): p. 1329-46. 46.

Harrington, K.J., S. Mohammadtaghi, P.S. Uster, D. Glass, A.M. Peters, R.G. Vile, and J.S. Stewart, Effective targeting of solid tumors in patients with locally advanced cancers

by radiolabeled pegylated liposomes. Clin Cancer Res, 2001. 7(2): p. 243-54. 47.

Wu, P.C., C.H. Su, F.Y. Cheng, J.C. Weng, J.H. Chen, T.L. Tsai, C.S. Yeh, W.C. Su, J.R. Hwu, Y. Tzeng, and D.B. Shieh, Modularly assembled magnetite nanoparticles enhance

in vivo targeting for magnetic resonance cancer imaging. Bioconjug Chem, 2008.

19(10): p. 1972-9. 48.

Zhu, D., X. Lu, P.A. Hardy, M. Leggas, and M. Jay, Nanotemplate-engineered

nanoparticles containing gadolinium for magnetic resonance imaging of tumors. Invest Radiol, 2008. 43(2): p. 129-40. 49.

Yu, Y.B., Fluorocarbon nanoparticles as multifunctional drug delivery vehicles. J Drug Target, 2006. 14(10): p. 663-9.

50.

Viglianti, B.L., S.A. Abraham, C.R. Michelich, P.S. Yarmolenko, J.R. MacFall, M.B. Bally, and M.W. Dewhirst, In vivo monitoring of tissue pharmacokinetics of

liposome/drug using MRI: illustration of targeted delivery. Magn Reson Med, 2004.

51(6): p. 1153-62. 51.

Viglianti, B.L., A.M. Ponce, C.R. Michelich, D. Yu, S.A. Abraham, L. Sanders, P.S. Yarmolenko, T. Schroeder, J.R. MacFall, D.P. Barboriak, O.M. Colvin, M.B. Bally, and M.W. Dewhirst, Chemodosimetry of in vivo tumor liposomal drug concentration using

MRI. Magn Reson Med, 2006. 56(5): p. 1011-8. 52.

Cassidy, P.J. and G.K. Radda, Molecular imaging perspectives. J R Soc Interface, 2005.

2(3): p. 133-44. 53.

Shah, N.J., M. Zaitsev, S. Steinhoff, and K. Zilles, A new method for fast multislice T(1)

mapping. Neuroimage, 2001. 14(5): p. 1175-85. 54.

Steinhoff, S., M. Zaitsev, K. Zilles, and N.J. Shah, Fast T(1) mapping with volume

coverage. Magn Reson Med, 2001. 46(1): p. 131-40.

130 55.

Shah, N.J., H. Neeb, M. Zaitsev, S. Steinhoff, G. Kircheis, K. Amunts, D. Haussinger, and K. Zilles, Quantitative T1 mapping of hepatic encephalopathy using magnetic

resonance imaging. Hepatology, 2003. 38(5): p. 1219-26. 56.

Deoni, S.C., H.A. Ward, T.M. Peters, and B.K. Rutt, Rapid T2 estimation with phase-

cycled variable nutation steady-state free precession. Magn Reson Med, 2004. 52(2): p. 435-9. 57.

Blamire, A.M., The technology of MRI--the next 10 years? Br J Radiol, 2008. 81(968): p. 601-17.

58.

Clare, S. and P. Jezzard, Rapid T(1) mapping using multislice echo planar imaging. Magn Reson Med, 2001. 45(4): p. 630-4.

59.

Lanao, J.M. and M.A. Fraile, Drug tissue distribution: study methods and therapeutic

implications. Curr Pharm Des, 2005. 11(29): p. 3829-45. 60.

Riggs, J.E.R. and R. Bachur, Clinical pharmacokinetics of anthracycline antibiotics, in

Pharmacokinetics of anticancer agents in humans, M.M. Ames, G. Powis, and J.S. Kovach, Editors. 1983, Elsevier: Amsterdam. p. 229-278. 61.

Zheng, J., G. Perkins, A. Kirilova, C. Allen, and D.A. Jaffray, Multimodal contrast agent

for combined computed tomography and magnetic resonance imaging applications. Invest Radiol, 2006. 41(3): p. 339-348. 62.

Zheng, J., J. Liu, M. Dunne, D.A. Jaffray, and C. Allen, In vivo performance of a

liposomal vascular contrast agent for CT and MR-based image guidance applications. Pharm Res, 2007. 24(6): p. 1193-201. 63.

Zheng, J., D.A. Jaffray, and C. Allen, Quantitative CT imaging of the spatial and

temporal distribution of liposomes in a rabbit tumor model. Mol Pharm, 2009. In press. 64.

Townsend, D.W. and T. Beyer, A combined PET/CT scanner: the path to true image

fusion. Br J Radiol, 2002. 75 Spec No: p. S24-30. 65.

Townsend, D.W., J.P. Carney, J.T. Yap, and N.C. Hall, PET/CT today and tomorrow. J Nucl Med, 2004. 45 Suppl 1: p. 4S-14S.

66.

Townsend, D.W., From 3-D positron emission tomography to 3-D positron emission

tomography/computed tomography: what did we learn? Mol Imaging Biol, 2004. 6(5): p. 275-90.

131 67.

Bocher, M., A. Balan, Y. Krausz, Y. Shrem, A. Lonn, M. Wilk, and R. Chisin, Gamma

camera-mounted anatomical X-ray tomography: technology, system characteristics and first images. Eur J Nucl Med, 2000. 27(6): p. 619-27. 68.

Patton, J.A., D. Delbeke, and M.P. Sandler, Image fusion using an integrated, dual-head

coincidence camera with X-ray tube-based attenuation maps. J Nucl Med, 2000. 41(8): p. 1364-8. 69.

Fahrig, R., G. Heit, Z. Wen, B.L. Daniel, K. Butts, and N.J. Pelc, First use of a truly-

hybrid X-ray/MR imaging system for guidance of brain biopsy. Acta Neurochir (Wien), 2003. 145(11): p. 995-7; discussion 997. 70.

Fahrig, R., K. Butts, Z. Wen, R. Saunders, S.T. Kee, D.Y. Sze, B.L. Daniel, F. Laerum, and N.J. Pelc, Truly hybrid interventional MR/X-ray system: investigation of in vivo

applications. Acad Radiol, 2001. 8(12): p. 1200-7. 71.

Fahrig, R., K. Butts, J.A. Rowlands, R. Saunders, J. Stanton, G.M. Stevens, B.L. Daniel, Z. Wen, D.L. Ergun, and N.J. Pelc, A truly hybrid interventional MR/X-ray system:

feasibility demonstration. J Magn Reson Imaging, 2001. 13(2): p. 294-300. 72.

Barillot, C., D. Lemoine, L. Le Briquer, F. Lachmann, and B. Gibaud, Data fusion in

medical imaging: merging multimodal and multipatient images, identification of structures and 3D display aspects. Eur J Radiol, 1993. 17(1): p. 22-7. 73.

Branstetter, B.F.t., T.M. Blodgett, L.A. Zimmer, C.H. Snyderman, J.T. Johnson, S. Raman, and C.C. Meltzer, Head and neck malignancy: is PET/CT more accurate than

PET or CT alone? Radiology, 2005. 235(2): p. 580-6. 74.

Bar-Shalom, R., L. Guralnik, M. Tsalic, M. Leiderman, A. Frenkel, D. Gaitini, A. BenNun, Z. Keidar, and O. Israel, The additional value of PET/CT over PET in FDG imaging

of oesophageal cancer. Eur J Nucl Med Mol Imaging, 2005. 75.

Coleman RE, Delbeke D, Guilberteau MJ, Conti PS, Royal HD, Weinreb JC, Siegel BA, Federle MF, Townsend DW, Berland LL., Concurrent PET/CT with and integrated

imaging system: intersociety dialogue from the Joint Working Group of the American College of Radiology, the Society of Nuclear Medicine, and the Society of Computed Body Tomography and Magnetic Resonance. J Nucl Med, 2005. 46(7): p. 1-15.

132 76.

Rosenman, J.G., E.P. Miller, G. Tracton, and T.J. Cullip, Image registration: an essential

part of radiation therapy treatment planning. Int J Radiat Oncol Biol Phys, 1998. 40(1): p. 197-205. 77.

Rasch, C., I. Barillot, P. Remeijer, A. Touw, M. van Herk, and J.V. Lebesque, Definition

of the prostate in CT and MRI: a multi-observer study. Int J Radiat Oncol Biol Phys, 1999. 43(1): p. 57-66. 78.

Uematsu, M., M. Sonderegger, A. Shioda, K. Tahara, T. Fukui, Y. Hama, T. Kojima, J.R. Wong, and S. Kusano, Daily positioning accuracy of frameless stereotactic radiation

therapy with a fusion of computed tomography and linear accelerator (focal) unit: evaluation of z-axis with a z-marker. Radiother Oncol, 1999. 50(3): p. 337-9. 79.

Jaffray, D.A., J.H. Siewerdsen, J.W. Wong, and A.A. Martinez, Flat-panel cone-beam

computed tomography for image-guided radiation therapy. Int J Radiat Oncol Biol Phys, 2002. 53(5): p. 1337-49. 80.

Krause, W., Delivery of diagnostic agents in computed tomography. Adv Drug Deliv Rev, 1999. 37(1-3): p. 159-173.

81.

Saeed, M., M.F. Wendland, and C.B. Higgins, Blood pool MR contrast agents for

cardiovascular imaging. J Magn Reson Imaging, 2000. 12(6): p. 890-8. 82.

Patty, P.J. and B.J. Frisken, The pressure-dependence of the size of extruded vesicles. Biophys J, 2003. 85(2): p. 996-1004.

83.

Hunter, D.G. and B.J. Frisken, Effect of extrusion pressure and lipid properties on the

size and polydispersity of lipid vesicles. Biophys J, 1998. 74(6): p. 2996-3002. 84.

Liu, J., Y. Xiao, and C. Allen, Polymer-drug compatibility: a guide to the development of

delivery systems for the anticancer agent, ellipticine. J Pharm Sci, 2004. 93(1): p. 132-43. 85.

Carr H, Purcell E., Effects of diffusion on free precession in nuclear magnetic resonance

experiments. Phys Rev, 1954. 94: p. 630-638. 86.

Meiboom S, Gill S., Modified spin-echo method for measuring nuclear relaxation times. Rev Sci Instrum., 1958. 29: p. 668-691.

87.

Takeda, M., Y. Katayama, T. Tsutsui, T. Komeyama, T. Mizusawa, T. Tanikawa, and S. Sato, Concentration of gadolinium-diethylene triamine pentaacetic acid in human kidney-

-study on proper time for dynamic magnetic resonance imaging of the human kidney on low and high magnetic fields. Tohoku J Exp Med, 1993. 171(2): p. 119-28.

133 88.

Tweedle, M.F., P. Wedeking, J. Telser, C.H. Sotak, C.A. Chang, K. Kumar, X. Wan, and S.M. Eaton, Dependence of MR signal intensity on Gd tissue concentration over a broad

dose range. Magn Reson Med, 1991. 22(2): p. 191-4; discussion 195-6. 89.

Morkenborg, J., M. Pedersen, F.T. Jensen, H. Stodkilde-Jorgensen, J.C. Djurhuus, and J. Frokiaer, Quantitative assessment of Gd-DTPA contrast agent from signal enhancement:

an in-vitro study. Magn Reson Imaging, 2003. 21(6): p. 637-43. 90.

Choyke, P.L., J.A. Frank, M.E. Girton, S.W. Inscoe, M.J. Carvlin, J.L. Black, H.A. Austin, and A.J. Dwyer, Dynamic Gd-DTPA-enhanced MR imaging of the kidney:

experimental results. Radiology, 1989. 170(3 Pt 1): p. 713-20. 91.

Carvlin, M.J., P.H. Arger, H.L. Kundel, L. Axel, L. Dougherty, E.A. Kassab, and B. Moore, Use of Gd-DTPA and fast gradient-echo and spin-echo MR imaging to

demonstrate renal function in the rabbit. Radiology, 1989. 170(3 Pt 1): p. 705-11. 92.

May, D.A. and D.J. Pennington, Effect of gadolinium concentration on renal signal

intensity: An in vitro study with a saline bag model. Radiology, 2000. 216(1): p. 232-6. 93.

Davis, P.L., D.L. Parker, J.A. Nelson, J.S. Gillen, and V.M. Runge, Interactions of

paramagnetic contrast agents and the spin echo pulse sequence. Invest Radiol, 1988.

23(5): p. 381-8. 94.

Montgomery, D.D., W.B. Morrison, M.E. Schweitzer, D. Weishaupt, and L. Dougherty,

Effects of iodinated contrast and field strength on gadolinium enhancement: implications for direct MR arthrography. J Magn Reson Imaging, 2002. 15(3): p. 334-43. 95.

Kopka, L., M. Funke, U. Fischer, D. Keating, J. Oestmann, and E. Grabbe, MR

arthrography of the shoulder with gadopentetate dimeglumine: influence of concentration, iodinated contrast material, and time on signal intensity. AJR Am J Roentgenol, 1994. 163(3): p. 621-3. 96.

Kopka, L., M. Funke, U. Fischer, R. Vosshenrich, M. Schroder, and E. Grabbe, [Signal

characteristics of x-ray contrast media and their interaction with gadolinium-DTPA in MRT]. Rofo, 1994. 160(4): p. 349-52. 97.

Fossheim, S.L., A.K. Fahlvik, J. Klaveness, and R.N. Muller, Paramagnetic liposomes as

MRI contrast agents: influence of liposomal physicochemical properties on the in vitro relaxivity. Magn Reson Imaging, 1999. 17(1): p. 83-9.

134 98.

Seibert, J.A., X-ray imaging physics for nuclear medicine technologists. Part 1: Basic

principles of x-ray production. J Nucl Med Technol, 2004. 32(3): p. 139-47. 99.

Kubicek, V.a.T.E., Design and function of metal complexes, in Advances in Inorganic

Chemistry, R.a.H. Van Eldik, C.D., Editor. 2009, Academic Press. 100.

McDonald MA, Watkin BS, Watkin KL., Small Particulate Gadolinium Oxide and

Gadolinium Oxide Albumin Microspheres as Multimodal Contrast and Therapeutic Agents. Invest Radiol, 2003. 38: p. 305-310. 101.

Bloem JL, Wondergem J., Gd-DTPA as a contrast agent in CT. Radiology., 1989. 171: p. 578-579.

102.

Gierda DS, Bae KT., Gadolinium as a CT contrast agent: assessment in a procine model. Radiology., 1999. 210: p. 829-834.

103.

Quinn AD, O'Hare NJ, Wallis FJ, et al., Gd-DTPA: an alternative contrast medium for

CT. J Comput Assist Tomogr., 1994. 18: p. 634-636. 104.

Pena CS, Kaufman JA, Geller SC, et al., Gadopentetate dimeglumine: a possible

alternative contrast agent for CT angiography of the aorta. J Comput Assist Tomogr., 1999. 23: p. 23-24. 105.

Kircher, M.F., U. Mahmood, R.S. King, R. Weissleder, and L. Josephson, A multimodal

nanoparticle for preoperative magnetic resonance imaging and intraoperative optical brain tumor delineation. Cancer Res, 2003. 63(23): p. 8122-5. 106.

Kao, C.Y., E.A. Hoffman, K.C. Beck, R.V. Bellamkonda, and A.V. Annapragada, Long-

residence-time nano-scale liposomal iohexol for X-ray-based blood pool imaging. Acad Radiol, 2003. 10(5): p. 475-83. 107.

Leike, J.U., A. Sachse, and K. Rupp, Characterization of continuously extruded

iopromide-carrying liposomes for computed tomography blood-pool imaging. Invest Radiol, 2001. 36(6): p. 303-8. 108.

Leander, P., P. Hoglund, A. Borseth, Y. Kloster, and A. Berg, A new liposomal liver-

specific contrast agent for CT: first human phase-I clinical trial assessing efficacy and safety. Eur Radiol, 2001. 11(4): p. 698-704. 109.

Schmiedl, U.P., W. Krause, J. Leike, and A. Sachse, CT blood pool enhancement in

primates with lopromide-carrying liposomes containing soy phosphatidyl glycerol. Acad Radiol, 1999. 6(3): p. 164-9.

135 110.

Spinazzi, A., S. Ceriati, P. Pianezzola, V. Lorusso, F. Luzzani, X. Fouillet, S. Alvino, and E.J. Rummeny, Safety and pharmacokinetics of a new liposomal liver-specific contrast

agent for CT: results of clinical testing in nonpatient volunteers. Invest Radiol, 2000.

35(1): p. 1-7. 111.

Petersein, J., B. Franke, X. Fouillet, and B. Hamm, Evaluation of liposomal contrast

agents for liver CT in healthy rabbits. Invest Radiol, 1999. 34(6): p. 401-9. 112.

Leander, P., P. Hoglund, Y. Kloster, and A. Borseth, New liposomal liver-specific

contrast agent for CT: first human phase I clinical trial assessing efficacy and safety. Acad Radiol, 1998. 5 Suppl 1: p. S6-8; discussion S28-30. 113.

Krause, W., J. Leike, G. Schuhmann-Giampieri, A. Sachse, U. Schmiedl, and H. Strunk,

Iopromide-carrying liposomes as a contrast agent for the liver. Acad Radiol, 1996. 3

Suppl 2: p. S235-7. 114.

Dick, A., G. Adam, J. Tacke, A. Prescher, T.E. Southon, and R.W. Gunther, Computed

tomography of experimental liver abscesses using a new liposomal contrast agent. Invest Radiol, 1996. 31(4): p. 194-203. 115.

Revel, D., C. Corot, Y. Carrillon, G. Dandis, R. Eloy, and M. Amiel, Ioxaglate-carrying

liposomes. Computed tomographic study as hepatosplenic contrast agent in rabbits. Invest Radiol, 1990. 25 Suppl 1: p. S95-7. 116.

Musu, C., E. Felder, B. Lamy, and M. Schneider, A liposomal contrast agent.

Preliminary communication. Invest Radiol, 1988. 23 Suppl 1: p. S126-9. 117.

Zalutsky, M.R., M.A. Noska, and S.E. Seltzer, Characterization of liposomes containing

iodine-125-labeled radiographic contrast agents. Invest Radiol, 1987. 22(2): p. 141-7. 118.

Seltzer, S.E., P.M. Shulkin, D.F. Adams, M.A. Davis, G.B. Hoey, R.M. Hopkins, and M.E. Bosworth, Usefulness of liposomes carrying losefamate for CT opacification of

liver and spleen. AJR Am J Roentgenol, 1984. 143(3): p. 575-9. 119.

Jendrasiak, G.L., G.D. Frey, and R.C. Heim, Jr., Liposomes as carriers of iodolipid

radiocontrast agents for CT scanning of the liver. Invest Radiol, 1985. 20(9): p. 9951002. 120.

Torchilin, V.P., Polymeric contrast agents for medical imaging. Curr Pharm Biotechnol, 2000. 1(2): p. 183-215.

136 121.

Schneider T, Sachse A, Roβling G, Brandl M., Generation of contrast-carrying

liposomes of defined size with a new continuous high pressure extrusion method. Int J Pharm, 1995. 117: p. 1-12. 122.

Pauser, S., R. Reszka, S. Wagner, K.J. Wolf, H.J. Buhr, and G. Berger, Liposome-

encapsulated superparamagnetic iron oxide particles as markers in an MRI-guided search for tumor-specific drug carriers. Anticancer Drug Des, 1997. 12(2): p. 125-35. 123.

Weissig, V.V., J. Babich, and V.V. Torchilin, Long-circulating gadolinium-loaded

liposomes: potential use for magnetic resonance imaging of the blood pool. Colloids Surf B Biointerfaces, 2000. 18(3-4): p. 293-299. 124.

Misselwitz, B. and A. Sachse, Interstitial MR lymphography using Gd-carrying

liposomes. Acta Radiol Suppl, 1997. 412: p. 51-5. 125.

Unger, E., P. Needleman, P. Cullis, and C. Tilcock, Gadolinium-DTPA liposomes as a

potential MRI contrast agent. Work in progress. Invest Radiol, 1988. 23(12): p. 928-32. 126.

Kabalka, G., E. Buonocore, K. Hubner, T. Moss, N. Norley, and L. Huang, Gadolinium-

labeled liposomes: targeted MR contrast agents for the liver and spleen. Radiology, 1987. 163(1): p. 255-8. 127.

Grant, C.W., S. Karlik, and E. Florio, A liposomal MRI contrast agent:

phosphatidylethanolamine-DTPA. Magn Reson Med, 1989. 11(2): p. 236-43. 128.

Raffy, S. and J. Teissie, Control of lipid membrane stability by cholesterol content. Biophys J, 1999. 76(4): p. 2072-80.

129.

Lasic, D.D., Novel applications of liposomes. Trends Biotechnol, 1998. 16(7): p. 307-21.

130.

Drummond, D.C., O. Meyer, K. Hong, D.B. Kirpotin, and D. Papahadjopoulos,

Optimizing liposomes for delivery of chemotherapeutic agents to solid tumors. Pharmacol Rev, 1999. 51(4): p. 691-743. 131.

Bloom, M., E. Evans, and O.G. Mouritsen, Physical properties of the fluid lipid-bilayer

component of cell membranes: a perspective. Q Rev Biophys, 1991. 24(3): p. 293-397. 132.

Moghimi, S.M. and H.M. Patel, Tissue specific opsonins for phagocytic cells and their

different affinity for cholesterol-rich liposomes. FEBS Lett, 1988. 233(1): p. 143-7. 133.

Patel, H.M., Serum opsonins and liposomes: their interaction and opsonophagocytosis. Crit Rev Ther Drug Carrier Syst, 1992. 9(1): p. 39-90.

137 134.

Patel, H.M., N.S. Tuzel, and B.E. Ryman, Inhibitory effect of cholesterol on the uptake of

liposomes by liver and spleen. Biochim Biophys Acta, 1983. 761(2): p. 142-51. 135.

Kirby, C. and G. Gregoriadis, The effect of the cholesterol content of small unilamellar

liposomes on the fate of their lipid components in vitro. Life Sci, 1980. 27(23): p. 222330. 136.

Allen, C., N. Dos Santos, R. Gallagher, G.N. Chiu, Y. Shu, W.M. Li, S.A. Johnstone, A.S. Janoff, L.D. Mayer, M.S. Webb, and M.B. Bally, Controlling the physical behavior

and biological performance of liposome formulations through use of surface grafted poly(ethylene glycol). Biosci Rep, 2002. 22(2): p. 225-50. 137.

Allen, T.M. and C. Hansen, Pharmacokinetics of stealth versus conventional liposomes:

effect of dose. Biochim Biophys Acta, 1991. 1068(2): p. 133-41. 138.

Trubetskoy, V.S., J.A. Cannillo, A. Milshtein, G.L. Wolf, and V.P. Torchilin, Controlled

delivery of Gd-containing liposomes to lymph nodes: surface modification may enhance MRI contrast properties. Magn Reson Imaging, 1995. 13(1): p. 31-7. 139.

Perkins GJ, Zheng J, Brock K, Allen C, Jaffray DA., Nanoengineered multimodal

contrast agent for medical image guidance. Proceedings of SPIE, 2005. 5746: p. 31-39. 140.

Pain, R.W., Body fluid compartments. Anaesth Intensive Care, 1977. 5(4): p. 284-94.

141.

Fritz, T., E. Unger, S. Wilson-Sanders, Q.F. Ahkong, and C. Tilcock, Detailed toxicity

studies of liposomal gadolinium-DTPA. Invest Radiol, 1991. 26(11): p. 960-8. 142.

Kalender, X-ray based imaging for computer-assisted surgery. Minim Invasive Ther Allied Technol, 2003. 12(1): p. 52-8.

143.

Sequeiros, R.B., R. Ojala, J. Kariniemi, J. Perala, J. Niinimaki, H. Reinikainen, and O. Tervonen, MR-guided interventional procedures: a review. Acta Radiol, 2005. 46(6): p. 576-86.

144.

Peters, T.M., Image-guidance for surgical procedures. Phys Med Biol, 2006. 51(14): p. R505-40.

145.

Rafferty, M.A., J.H. Siewerdsen, Y. Chan, M.J. Daly, D.J. Moseley, D.A. Jaffray, and J.C. Irish, Intraoperative cone-beam CT for guidance of temporal bone surgery. Otolaryngol Head Neck Surg, 2006. 134(5): p. 801-8.

146.

Exadaktylos, A.K., J. Duwe, F. Eckstein, C. Stoupis, H. Schoenfeld, H. Zimmermann, and T.P. Carrel, The role of contrast-enhanced spiral CT imaging versus chest X-rays in

138

surgical therapeutic concepts and thoracic aortic injury: a 29-year Swiss retrospective analysis of aortic surgery. Cardiovasc J S Afr, 2005. 16(3): p. 162-5. 147.

Saeed, M., D. Saloner, O. Weber, A. Martin, C. Henk, and C. Higgins, MRI in guiding

and assessing intramyocardial therapy. Eur Radiol, 2005. 15(5): p. 851-63. 148.

Erdogan, S., A. Roby, R. Sawant, J. Hurley, and V.P. Torchilin, Gadolinium-loaded

polychelating polymer-containing cancer cell-specific immunoliposomes. J Liposome Res, 2006. 16(1): p. 45-55. 149.

Pirollo, K.F., J. Dagata, P. Wang, M. Freedman, A. Vladar, S. Fricke, L. Ileva, Q. Zhou, and E.H. Chang, A tumor-targeted nanodelivery system to improve early MRI detection

of cancer. Mol Imaging, 2006. 5(1): p. 41-52. 150.

Mulder, W.J., G.J. Strijkers, G.A. van Tilborg, A.W. Griffioen, and K. Nicolay, Lipid-

based nanoparticles for contrast-enhanced MRI and molecular imaging. NMR Biomed, 2006. 19(1): p. 142-64. 151.

Mukundan, S., Jr., K.B. Ghaghada, C.T. Badea, C.Y. Kao, L.W. Hedlund, J.M. Provenzale, G.A. Johnson, E. Chen, R.V. Bellamkonda, and A. Annapragada, A

liposomal nanoscale contrast agent for preclinical CT in mice. AJR Am J Roentgenol, 2006. 186(2): p. 300-7. 152.

Martina, M.S., J.P. Fortin, C. Menager, O. Clement, G. Barratt, C. Grabielle-Madelmont, F. Gazeau, V. Cabuil, and S. Lesieur, Generation of superparamagnetic liposomes

revealed as highly efficient MRI contrast agents for in vivo imaging. J Am Chem Soc, 2005. 127(30): p. 10676-85. 153.

Moran, C.M., J.A. Ross, C. Cunningham, M. Butler, T. Anderson, D. Newby, K.A. Fox, and W.N. McDicken, Manufacture and acoustical characterisation of a high-frequency

contrast agent for targeting applications. Ultrasound Med Biol, 2006. 32(3): p. 421-8. 154.

Krauze, M.T., J. Forsayeth, J.W. Park, and K.S. Bankiewicz, Real-time Imaging and

Quantification of Brain Delivery of Liposomes. Pharm Res, 2006. 155.

Vaccaro, M., A. Accardo, D. Tesauro, G. Mangiapia, D. Lof, K. Schillen, O. Soderman, G. Morelli, and L. Paduano, Supramolecular Aggregates of Amphiphilic Gadolinium

Complexes

as

Blood

Pool

MRI/MRA

Contrast

Characterization. Langmuir, 2006. 22(15): p. 6635-6643.

Agents:

Physicochemical

139 156.

Lee, H.Y., H.W. Jee, S.M. Seo, B.K. Kwak, G. Khang, and S.H. Cho,

Diethylenetriaminepentaacetic acid-gadolinium (DTPA-Gd)-conjugated polysuccinimide derivatives as magnetic resonance imaging contrast agents. Bioconjug Chem, 2006.

17(3): p. 700-6. 157.

Accardo, A., D. Tesauro, P. Roscigno, E. Gianolio, L. Paduano, G. D'Errico, C. Pedone, and G. Morelli, Physicochemical properties of mixed micellar aggregates containing

CCK peptides and Gd complexes designed as tumor specific contrast agents in MRI. J Am Chem Soc, 2004. 126(10): p. 3097-107. 158.

Torchilin, V.P., PEG-based micelles as carriers of contrast agents for different imaging

modalities. Adv Drug Deliv Rev, 2002. 54(2): p. 235-52. 159.

Torchilin, V.P., M.D. Frank-Kamenetsky, and G.L. Wolf, CT visualization of blood pool

in rats by using long-circulating, iodine-containing micelles. Acad Radiol, 1999. 6(1): p. 61-5. 160.

Zhu, D., R.D. White, P.A. Hardy, N. Weerapreeyakul, K. Sutthanut, and M. Jay,

Biocompatible nanotemplate-engineered nanoparticles containing gadolinium: stability and relaxivity of a potential MRI contrast agent. J Nanosci Nanotechnol, 2006. 6(4): p. 996-1003. 161.

Corbin, I.R., H. Li, J. Chen, S. Lund-Katz, R. Zhou, J.D. Glickson, and G. Zheng, Low-

density lipoprotein nanoparticles as magnetic resonance imaging contrast agents. Neoplasia, 2006. 8(6): p. 488-98. 162.

McDonald, M.A. and K.L. Watkin, Investigations into the physicochemical properties of

dextran small particulate gadolinium oxide nanoparticles. Acad Radiol, 2006. 13(4): p. 421-7. 163.

Lee, H., E. Lee, K. Kim do, N.K. Jang, Y.Y. Jeong, and S. Jon, Antibiofouling polymer-

coated superparamagnetic iron oxide nanoparticles as potential magnetic resonance contrast agents for in vivo cancer imaging. J Am Chem Soc, 2006. 128(22): p. 7383-9. 164.

Hainfeld, J.F., D.N. Slatkin, T.M. Focella, and H.M. Smilowitz, Gold nanoparticles: a

new X-ray contrast agent. Br J Radiol, 2006. 79(939): p. 248-53. 165.

Rabin, O., J. Manuel Perez, J. Grimm, G. Wojtkiewicz, and R. Weissleder, An X-ray

computed tomography imaging agent based on long-circulating bismuth sulphide nanoparticles. Nat Mater, 2006. 5(2): p. 118-22.

140 166.

Talanov, V.S., C.A. Regino, H. Kobayashi, M. Bernardo, P.L. Choyke, and M.W. Brechbiel, Dendrimer-based nanoprobe for dual modality magnetic resonance and

fluorescence imaging. Nano Lett, 2006. 6(7): p. 1459-63. 167.

Kobayashi, H. and M.W. Brechbiel, Nano-sized MRI contrast agents with dendrimer

cores. Adv Drug Deliv Rev, 2005. 57(15): p. 2271-86. 168.

Langereis, S., Q.G. de Lussanet, M.H. van Genderen, E.W. Meijer, R.G. Beets-Tan, A.W. Griffioen, J.M. van Engelshoven, and W.H. Backes, Evaluation of Gd(III)DTPA-

terminated poly(propylene imine) dendrimers as contrast agents for MR imaging. NMR Biomed, 2006. 19(1): p. 133-41. 169.

Fu, Y., D.E. Nitecki, D. Maltby, G.H. Simon, K. Berejnoi, H.J. Raatschen, B.M. Yeh, D.M. Shames, and R.C. Brasch, Dendritic Iodinated Contrast Agents with PEG-Cores for

CT Imaging: Synthesis and Preliminary Characterization. Bioconjug Chem, 2006. 17(4): p. 1043-56. 170.

Port, M., C. Corot, X. Violas, P. Robert, I. Raynal, and G. Gagneur, How to compare the

efficiency of albumin-bound and nonalbumin-bound contrast agents in vivo: the concept of dynamic relaxivity. Invest Radiol, 2005. 40(9): p. 565-73. 171.

Zhang, Y., P.L. Choyke, H. Lu, H. Takahashi, R.B. Mannon, X. Zhang, H. Marcos, K.C. Li, and J.B. Kopp, Detection and localization of proteinuria by dynamic contrast-

enhanced magnetic resonance imaging using MS325. J Am Soc Nephrol, 2005. 16(6): p. 1752-7. 172.

Morawski, A.M., G.A. Lanza, and S.A. Wickline, Targeted contrast agents for magnetic

resonance imaging and ultrasound. Curr Opin Biotechnol, 2005. 16(1): p. 89-92. 173.

McDonald, M.A. and K.L. Watkin, Small particulate gadolinium oxide and gadolinium

oxide albumin microspheres as multimodal contrast and therapeutic agents. Invest Radiol, 2003. 38(6): p. 305-10. 174.

G.R. Norman, D.L. Streiner, Biostatistics: The Bare Essentials. 2000: B.C. Decker Inc.

175.

Mutzel, W. and U. Speck, Pharmacokinetics and biotransformation of iohexol in the rat

and the dog. Acta Radiol Suppl, 1980. 362: p. 87-92. 176.

Tweedle, M.F., X. Zhang, M. Fernandez, P. Wedeking, A.D. Nunn, and H.W. Strauss, A

noninvasive method for monitoring renal status at bedside. Invest Radiol, 1997. 32(12): p. 802-5.

141 177.

Heilmann, M., F. Kiessling, M. Enderlin, and L.R. Schad, Determination of

pharmacokinetic parameters in DCE MRI: Consequence of nonlinearity between contrast agent concentration and signal intensity. Invest Radiol, 2006. 41(6): p. 536-43. 178.

McLachlan, S.J., S. Eaton, and D.N. De Simone, Pharmacokinetic behavior of

gadoteridol injection. Invest Radiol, 1992. 27 Suppl 1: p. S12-5. 179.

Arvidsson, A. and A. Hedman, Plasma and renal clearance of iohexol--a study on the

reproducibility of a method for the glomerular filtration rate. Scand J Clin Lab Invest, 1990. 50(7): p. 757-61. 180.

Tweedle, M.F., The ProHance story: the making of a novel MRI contrast agent. Eur Radiol, 1997. 7 Suppl 5: p. 225-30.

181.

Kauczor, H.U., Multimodal imaging and computer assisted diagnosis for functional

tumour characterisation. Cancer Imaging, 2005. 5(1): p. 46-50. 182.

Choyke, P.L., Contrast agents for imaging tumor angiogenesis: is bigger better? Radiology, 2005. 235(1): p. 1-2.

183.

Dreher, M.R., W. Liu, C.R. Michelich, M.W. Dewhirst, F. Yuan, and A. Chilkoti, Tumor

vascular permeability, accumulation, and penetration of macromolecular drug carriers. J Natl Cancer Inst, 2006. 98(5): p. 335-44. 184.

Miles, K.A., Functional computed tomography in oncology. Eur J Cancer, 2002. 38(16): p. 2079-84.

185.

Daldrup-Link, H.E., G.H. Simon, and R.C. Brasch, Imaging of tumor angiogenesis:

current approaches and future prospects. Curr Pharm Des, 2006. 12(21): p. 2661-72. 186.

Allen, T.M., W.W. Cheng, J.I. Hare, and K.M. Laginha, Pharmacokinetics and

pharmacodynamics of lipidic nano-particles in cancer. Anticancer Agents Med Chem, 2006. 6(6): p. 513-23. 187.

Gabizon, A., D. Tzemach, L. Mak, M. Bronstein, and A.T. Horowitz, Dose dependency

of pharmacokinetics and therapeutic efficacy of pegylated liposomal doxorubicin (DOXIL) in murine models. J Drug Target, 2002. 10(7): p. 539-48. 188.

Cui, J., C. Li, W. Guo, Y. Li, C. Wang, L. Zhang, Y. Hao, and Y. Wang, Direct

comparison of two pegylated liposomal doxorubicin formulations: is AUC predictive for toxicity and efficacy? J Control Release, 2007. 118(2): p. 204-15.

142 189.

Ishida, T., K. Atobe, X. Wang, and H. Kiwada, Accelerated blood clearance of

PEGylated liposomes upon repeated injections: effect of doxorubicin-encapsulation and high-dose first injection. J Control Release, 2006. 115(3): p. 251-8. 190.

Lu, Z.R., F. Ye, and A. Vaidya, Polymer platforms for drug delivery and biomedical

imaging. J Control Release, 2007. 122(3): p. 269-77. 191.

Koning, G.A. and G.C. Krijger, Targeted multifunctional lipid-based nanocarriers for

image-guided drug delivery. Anticancer Agents Med Chem, 2007. 7(4): p. 425-40. 192.

Tashjian, J.A., M.W. Dewhirst, D. Needham, and B.L. Viglianti, Rationale for and

measurement of liposomal drug delivery with hyperthermia using non-invasive imaging techniques. Int J Hyperthermia, 2008. 24(1): p. 79-90. 193.

Nasongkla, N., E. Bey, J. Ren, H. Ai, C. Khemtong, J.S. Guthi, S.F. Chin, A.D. Sherry, D.A. Boothman, and J. Gao, Multifunctional polymeric micelles as cancer-targeted, MRI-

ultrasensitive drug delivery systems. Nano Lett, 2006. 6(11): p. 2427-30. 194.

Mitra, A., A. Nan, B.R. Line, and H. Ghandehari, Nanocarriers for nuclear imaging and

radiotherapy of cancer. Curr Pharm Des, 2006. 12(36): p. 4729-49. 195.

Torchilin, V.P., Targeted pharmaceutical nanocarriers for cancer therapy and imaging. Aaps J, 2007. 9(2): p. E128-47.

196.

Hoffman, A.S., The origins and evolution of "controlled" drug delivery systems. J Control Release, 2008.

197.

Greish, K., Enhanced permeability and retention of macromolecular drugs in solid

tumors: a royal gate for targeted anticancer nanomedicines. J Drug Target, 2007. 15(78): p. 457-64. 198.

Torchilin, V.P., Micellar nanocarriers: pharmaceutical perspectives. Pharm Res, 2007.

24(1): p. 1-16. 199.

Drummond, D.C., C.O. Noble, M.E. Hayes, J.W. Park, and D.B. Kirpotin,

Pharmacokinetics and in vivo drug release rates in liposomal nanocarrier development. J Pharm Sci, 2008. 97(11): p. 4696-740. 200.

Vail, D.M., M.A. Amantea, G.T. Colbern, F.J. Martin, R.A. Hilger, and P.K. Working,

Pegylated liposomal doxorubicin: proof of principle using preclinical animal models and pharmacokinetic studies. Semin Oncol, 2004. 31(6 Suppl 13): p. 16-35.

143 201.

Goins, B.A. and W.T. Phillips, The use of scintigraphic imaging as a tool in the

development of liposome formulations. Prog Lipid Res, 2001. 40(1-2): p. 95-123. 202.

Charrois, G.J. and T.M. Allen, Drug release rate influences the pharmacokinetics,

biodistribution, therapeutic activity, and toxicity of pegylated liposomal doxorubicin formulations in murine breast cancer. Biochim Biophys Acta, 2004. 1663(1-2): p. 16777. 203.

Zheng, J., J.D. Hoisak, C. Allen, and D.A. Jaffray, Longitudinal vascular imaging using a

novel nano-encapsulated CT and MR contrast agent. Proc. of SPIE, 2007. 6511: p. D1D9. 204.

Pathak, A.P., Magnetic resonance imaging of tumor physiology. Methods Mol Med, 2006. 124: p. 279-97.

205.

Brock, K.K., L.A. Dawson, M.B. Sharpe, D.J. Moseley, and D.A. Jaffray, Feasibility of a

novel deformable image registration technique to facilitate classification, targeting, and monitoring of tumor and normal tissue. Int J Radiat Oncol Biol Phys, 2006. 64(4): p. 1245-54. 206.

Hutton, B.F. and M. Braun, Software for image registration: algorithms, accuracy,

efficacy. Semin Nucl Med, 2003. 33(3): p. 180-92. 207.

Ford, N.L., K.C. Graham, A.C. Groom, I.C. Macdonald, A.F. Chambers, and D.W. Holdsworth, Time-course characterization of the computed tomography contrast

enhancement of an iodinated blood-pool contrast agent in mice using a volumetric flatpanel equipped computed tomography scanner. Invest Radiol, 2006. 41(4): p. 384-90. 208.

Zerhouni, E., Medicine. The NIH Roadmap. Science, 2003. 302(5642): p. 63-72.

209.

Weis, S.M., Vascular permeability in cardiovascular disease and cancer. Curr Opin Hematol, 2008. 15(3): p. 243-9.

210.

Dams, E.T., W.J. Oyen, O.C. Boerman, G. Storm, P. Laverman, P.J. Kok, W.C. Buijs, H. Bakker, J.W. van der Meer, and F.H. Corstens, 99mTc-PEG liposomes for the

scintigraphic detection of infection and inflammation: clinical evaluation. J Nucl Med, 2000. 41(4): p. 622-30. 211.

Lammers, T., W.E. Hennink, and G. Storm, Tumour-targeted nanomedicines: principles

and practice. Br J Cancer, 2008. 99(3): p. 392-7.

144 212.

Gabizon, A.A., Pegylated liposomal doxorubicin: metamorphosis of an old drug into a

new form of chemotherapy. Cancer Invest, 2001. 19(4): p. 424-36. 213.

de Crespigny, A.J., D. Howard, H. D'Arceuil, H. Muller, A.T. Agoston, S. Seri, Y. Hashiguchi, C. Fujimoto, A. Nakatani, and M.E. Moseley, Dynamic contrast-enhanced

MRI of Implanted VX2 tumors in rabbit muscle: comparison of Gd-DTPA and NMS60. Magn Reson Imaging, 1999. 17(9): p. 1297-305. 214.

Park, K.S., B.I. Choi, H.J. Won, J.B. Seo, S.H. Kim, T.K. Kim, J.K. Han, and K.M. Yeon, Intratumoral vascularity of experimentally induced VX2 carcinoma: comparison of

color Doppler sonography, power Doppler sonography, and microangiography. Invest Radiol, 1998. 33(1): p. 39-44. 215.

Yamada, K., T. Jinbo, K. Miyahara, M. Sato, T. Hirose, H. Kato, Y. Tateno, H. Ikehira, H. Sugihara, and K. Furuhama, Contrast-enhanced MRI with gadodiamide injection in

rabbit carcinoma models. J Vet Med Sci, 1996. 58(5): p. 389-96. 216.

Choi, B.I., D.H. Lee, and M.C. Han, Necrotic areas in VX2 carcinoma of rabbits.

Correlation of magnetic resonance imaging and pathologic appearance. Invest Radiol, 1993. 28(1): p. 33-8. 217.

Nakamoto, Y., B.B. Chin, D.L. Kraitchman, L.P. Lawler, L.T. Marshall, and R.L. Wahl,

Effects of nonionic intravenous contrast agents at PET/CT imaging: phantom and canine studies. Radiology, 2003. 227(3): p. 817-24. 218.

Soret, M., S.L. Bacharach, and I. Buvat, Partial-volume effect in PET tumor imaging. J Nucl Med, 2007. 48(6): p. 932-45.

219.

Silverman, P.M., Liver metastases: imaging considerations for protocol development

with multislice CT (MSCT). Cancer Imaging, 2006. 6: p. 175-81. 220.

Torchilin, V.P., Recent advances with liposomes as pharmaceutical carriers. Nat Rev Drug Discov, 2005. 4(2): p. 145-60.

221.

Laverman, P., O.C. Boerman, W.J. Oyen, E.T. Dams, G. Storm, and F.H. Corstens,

Liposomes for scintigraphic detection of infection and inflammation. Adv Drug Deliv Rev, 1999. 37(1-3): p. 225-235. 222.

Laverman, P., E.T. Dams, G. Storm, T.G. Hafmans, H.J. Croes, W.J. Oyen, F.H. Corstens, and O.C. Boerman, Microscopic localization of PEG-liposomes in a rat model

of focal infection. J Control Release, 2001. 75(3): p. 347-55.

145 223.

Riedl, C.C., T. Akhurst, S. Larson, S.F. Stanziale, S. Tuorto, A. Bhargava, H. Hricak, D. Klimstra, and Y. Fong, 18F-FDG PET scanning correlates with tissue markers of poor

prognosis and predicts mortality for patients after liver resection for colorectal metastases. J Nucl Med, 2007. 48(5): p. 771-5. 224.

Guo, H., H. Zhu, Y. Xi, B. Zhang, L. Li, Y. Huang, J. Zhang, Z. Fu, G. Yang, S. Yuan, and J. Yu, Diagnostic and prognostic value of 18F-FDG PET/CT for patients with

suspected recurrence from squamous cell carcinoma of the esophagus. J Nucl Med, 2007. 48(8): p. 1251-8. 225.

Hamazawa, Y., K. Koyama, T. Okamura, Y. Wada, T. Wakasa, T. Okuma, Y. Watanabe, and Y. Inoue, Comparison of dynamic FDG-microPET study in a rabbit turpentine-

induced inflammatory model and in a rabbit VX2 tumor model. Ann Nucl Med, 2007.

21(1): p. 47-55. 226.

Ogihara, I., S. Kojima, and M. Jay, Differential uptake of gallium-67-labeled liposomes

between tumors and inflammatory lesions in rats. J Nucl Med, 1986. 27(8): p. 1300-7. 227.

Kumar, R., V.A. Loving, A. Chauhan, H. Zhuang, S. Mitchell, and A. Alavi, Potential of

dual-time-point imaging to improve breast cancer diagnosis with (18)F-FDG PET. J Nucl Med, 2005. 46(11): p. 1819-24. 228.

Hustinx, R., R.J. Smith, F. Benard, D.I. Rosenthal, M. Machtay, L.A. Farber, and A. Alavi, Dual time point fluorine-18 fluorodeoxyglucose positron emission tomography: a

potential method to differentiate malignancy from inflammation and normal tissue in the head and neck. Eur J Nucl Med, 1999. 26(10): p. 1345-8. 229.

Kubota, R., K. Kubota, S. Yamada, M. Tada, T. Ido, and N. Tamahashi,

Microautoradiographic study for the differentiation of intratumoral macrophages, granulation tissues and cancer cells by the dynamics of fluorine-18-fluorodeoxyglucose uptake. J Nucl Med, 1994. 35(1): p. 104-12. 230.

Kubota, R., S. Yamada, K. Kubota, K. Ishiwata, N. Tamahashi, and T. Ido, Intratumoral

distribution

of

fluorine-18-fluorodeoxyglucose

in

vivo:

high

accumulation

in

macrophages and granulation tissues studied by microautoradiography. J Nucl Med, 1992. 33(11): p. 1972-80.

146 231.

Kaim, A.H., B. Weber, M.O. Kurrer, J. Gottschalk, G.K. Von Schulthess, and A. Buck,

Autoradiographic quantification of 18F-FDG uptake in experimental soft-tissue abscesses in rats. Radiology, 2002. 223(2): p. 446-51. 232.

Huang, S.K., K.D. Lee, K. Hong, D.S. Friend, and D. Papahadjopoulos, Microscopic

localization of sterically stabilized liposomes in colon carcinoma-bearing mice. Cancer Res, 1992. 52(19): p. 5135-43. 233.

Sessa, G. and G. Weissmann, Phospholipid spherules (liposomes) as a model for

biological membranes. J Lipid Res, 1968. 9(3): p. 310-8. 234.

Caride, V.J., W. Taylor, J.A. Cramer, and A. Gottschalk, Evaluation of liposome-

entrapped radioactive tracers as scanning agents. Part 1: Organ distribution of liposome (99mTc-DTPA) in mice. J Nucl Med, 1976. 17(12): p. 1067-72. 235.

Caride, V.J., Technical and biological considerations on the use of radiolabeled

liposomes for diagnostic imaging. Int J Rad Appl Instrum B, 1990. 17(1): p. 35-9. 236.

Havron, A., S.E. Seltzer, M.A. Davis, and P. Shulkin, Radiopaque liposomes: a

promising new contrast material for computed tomography of the spleen. Radiology, 1981. 140(2): p. 507-11. 237.

Magin, R.L., S.M. Wright, M.R. Niesman, H.C. Chan, and H.M. Swartz, Liposome

delivery of NMR contrast agents for improved tissue imaging. Magn Reson Med, 1986.

3(3): p. 440-7. 238.

Unger, E.C., P.J. Lund, D.K. Shen, T.A. Fritz, D. Yellowhair, and T.E. New, Nitrogen-

filled liposomes as a vascular US contrast agent: preliminary evaluation. Radiology, 1992. 185(2): p. 453-6. 239.

Mulder, W.J., G.J. Strijkers, J.W. Habets, E.J. Bleeker, D.W. van der Schaft, G. Storm, G.A. Koning, A.W. Griffioen, and K. Nicolay, MR molecular imaging and fluorescence

microscopy for identification of activated tumor endothelium using a bimodal lipidic nanoparticle. Faseb J, 2005. 19(14): p. 2008-10. 240.

Zielhuis, S.W., J.H. Seppenwoolde, V.A. Mateus, C.J. Bakker, G.C. Krijger, G. Storm, B.A. Zonnenberg, A.D. van het Schip, G.A. Koning, and J.F. Nijsen, Lanthanide-loaded

liposomes for multimodality imaging and therapy. Cancer Biother Radiopharm, 2006.

21(5): p. 520-7.

147 241.

Josephson, L., M.F. Kircher, U. Mahmood, Y. Tang, and R. Weissleder, Near-infrared

fluorescent nanoparticles as combined MR/optical imaging probes. Bioconjug Chem, 2002. 13(3): p. 554-60. 242.

Schellenberger, E.A., D. Sosnovik, R. Weissleder, and L. Josephson, Magneto/optical

annexin V, a multimodal protein. Bioconjug Chem, 2004. 15(5): p. 1062-7. 243.

Sosnovik, D.E., E.A. Schellenberger, M. Nahrendorf, M.S. Novikov, T. Matsui, G. Dai, F. Reynolds, L. Grazette, A. Rosenzweig, R. Weissleder, and L. Josephson, Magnetic

resonance imaging of cardiomyocyte apoptosis with a novel magneto-optical nanoparticle. Magn Reson Med, 2005. 54(3): p. 718-24. 244.

Huh, Y.M., Y.W. Jun, H.T. Song, S. Kim, J.S. Choi, J.H. Lee, S. Yoon, K.S. Kim, J.S. Shin, J.S. Suh, and J. Cheon, In vivo magnetic resonance detection of cancer by using

multifunctional magnetic nanocrystals. J Am Chem Soc, 2005. 127(35): p. 12387-91. 245.

Cormode, D.P., T. Skajaa, M.M. van Schooneveld, R. Koole, P. Jarzyna, M.E. Lobatto, C. Calcagno, A. Barazza, R.E. Gordon, P. Zanzonico, E.A. Fisher, Z.A. Fayad, and W.J. Mulder, Nanocrystal core high-density lipoproteins: a multimodality contrast agent

platform. Nano Lett, 2008. 8(11): p. 3715-23. 246.

Koyama, Y., V.S. Talanov, M. Bernardo, Y. Hama, C.A. Regino, M.W. Brechbiel, P.L. Choyke, and H. Kobayashi, A dendrimer-based nanosized contrast agent dual-labeled for

magnetic resonance and optical fluorescence imaging to localize the sentinel lymph node in mice. J Magn Reson Imaging, 2007. 25(4): p. 866-71. 247.

McVeigh, E.R., M.A. Guttman, R.J. Lederman, M. Li, O. Kocaturk, T. Hunt, S. Kozlov, and K.A. Horvath, Real-time interactive MRI-guided cardiac surgery: aortic valve

replacement using a direct apical approach. Magn Reson Med, 2006. 56(5): p. 958-64. 248.

Hashizume, M., MRI-guided laparoscopic and robotic surgery for malignancies. Int J Clin Oncol, 2007. 12(2): p. 94-8.

249.

Rygh, O.M., T.A. Nagelhus Hernes, F. Lindseth, T. Selbekk, T. Brostrup Muller, and G. Unsgaard, Intraoperative navigated 3-dimensional ultrasound angiography in tumor

surgery. Surg Neurol, 2006. 66(6): p. 581-92; discussion 592. 250.

Lindner, D., C. Trantakis, C. Renner, S. Arnold, A. Schmitgen, J. Schneider, and J. Meixensberger, Application of intraoperative 3D ultrasound during navigated tumor

resection. Minim Invasive Neurosurg, 2006. 49(4): p. 197-202.

148 251.

Hsu, C.P., M.K. Razavi, S.K. So, I.H. Parachikov, and D.A. Benaron, Liver tumor gross

margin identification and ablation monitoring during liver radiofrequency treatment. J Vasc Interv Radiol, 2005. 16(11): p. 1473-8. 252.

Schmidt,

K.C.

and

C.B.

Smith,

Resolution,

sensitivity

and

precision

with

autoradiography and small animal positron emission tomography: implications for functional brain imaging in animal research. Nucl Med Biol, 2005. 32(7): p. 719-25. 253.

Diaspro, A., P. Bianchini, G. Vicidomini, M. Faretta, P. Ramoino, and C. Usai, Multi-

photon excitation microscopy. Biomed Eng Online, 2006. 5: p. 36. 254.

du Bois d'Aische, A., M.D. Craene, X. Geets, V. Gregoire, B. Macq, and S.K. Warfield,

Efficient multi-modal dense field non-rigid registration: alignment of histological and section images. Med Image Anal, 2005. 9(6): p. 538-46. 255.

Carlson, S.K., K.L. Classic, C.E. Bender, and S.J. Russell, Small animal absorbed

radiation dose from serial micro-computed tomography imaging. Mol Imaging Biol, 2007. 9(2): p. 78-82. 256.

Adams, A., C. SL, K. JC, H. RW, and Z. KR, Bioluminescence imaging confirms that

weekly computed tomography studies do not change tumor growth in an animal model of breast cancer metastasis. Mol Imaging Biol, 2006. 8(2): p. 76-77. 257.

Figueroa, S.D., C.T. Winkelmann, H.W. Miller, W.A. Volkert, and T.J. Hoffman, TLD

assessment of mouse dosimetry during microCT imaging. Med Phys, 2008. 35(9): p. 3866-74. 258.

Funk, T., M. Sun, and B.H. Hasegawa, Radiation dose estimate in small animal SPECT

and PET. Med Phys, 2004. 31(9): p. 2680-6. 259.

Bartlett, D.W., H. Su, I.J. Hildebrandt, W.A. Weber, and M.E. Davis, Impact of tumor-

specific targeting on the biodistribution and efficacy of siRNA nanoparticles measured by multimodality in vivo imaging. Proc Natl Acad Sci U S A, 2007. 104(39): p. 15549-54. 260.

Kirpotin, D.B., D.C. Drummond, Y. Shao, M.R. Shalaby, K. Hong, U.B. Nielsen, J.D. Marks, C.C. Benz, and J.W. Park, Antibody targeting of long-circulating lipidic

nanoparticles does not increase tumor localization but does increase internalization in animal models. Cancer Res, 2006. 66(13): p. 6732-40. 261.

Wu, A.M., P.J. Yazaki, S. Tsai, K. Nguyen, A.L. Anderson, D.W. McCarthy, M.J. Welch, J.E. Shively, L.E. Williams, A.A. Raubitschek, J.Y. Wong, T. Toyokuni, M.E.

149 Phelps, and S.S. Gambhir, High-resolution microPET imaging of carcinoembryonic

antigen-positive xenografts by using a copper-64-labeled engineered antibody fragment. Proc Natl Acad Sci U S A, 2000. 97(15): p. 8495-500. 262.

Sundaresan, G., P.J. Yazaki, J.E. Shively, R.D. Finn, S.M. Larson, A.A. Raubitschek, L.E. Williams, A.F. Chatziioannou, S.S. Gambhir, and A.M. Wu, 124I-labeled

engineered anti-CEA minibodies and diabodies allow high-contrast, antigen-specific small-animal PET imaging of xenografts in athymic mice. J Nucl Med, 2003. 44(12): p. 1962-9. 263.

Kircheis, R., L. Wightman, A. Schreiber, B. Robitza, V. Rossler, M. Kursa, and E. Wagner, Polyethylenimine/DNA complexes shielded by transferrin target gene expression

to tumors after systemic application. Gene Ther, 2001. 8(1): p. 28-40. 264.

Yu, W., K.F. Pirollo, B. Yu, A. Rait, L. Xiang, W. Huang, Q. Zhou, G. Ertem, and E.H. Chang, Enhanced transfection efficiency of a systemically delivered tumor-targeting

immunolipoplex by inclusion of a pH-sensitive histidylated oligolysine peptide. Nucleic Acids Res, 2004. 32(5): p. e48. 265.

Cheng, J., B.A. Teply, I. Sherifi, J. Sung, G. Luther, F.X. Gu, E. Levy-Nissenbaum, A.F. Radovic-Moreno, R. Langer, and O.C. Farokhzad, Formulation of functionalized PLGA-

PEG nanoparticles for in vivo targeted drug delivery. Biomaterials, 2007. 28(5): p. 86976. 266.

Pirollo, K.F. and E.H. Chang, Does a targeting ligand influence nanoparticle tumor

localization or uptake? Trends Biotechnol, 2008. 26(10): p. 552-8. 267.

Cho, H., E. Ackerstaff, S. Carlin, M.E. Lupu, Y. Wang, A. Rizwan, J. O'Donoghue, C.C. Ling, J.L. Humm, P.B. Zanzonico, and J.A. Koutcher, Noninvasive multimodality

imaging of the tumor microenvironment: registered dynamic magnetic resonance imaging and positron emission tomography studies of a preclinical tumor model of tumor hypoxia. Neoplasia, 2009. 11(3): p. 247-59, 2p following 259. 268.

Vaupel, P.W., The influence of tumor blood flow and microenvironmental factors on the

efficacy of radiation, drugs and localized hyperthermia. Klin Padiatr, 1997. 209(4): p. 243-9.