Glioblastoma multiforme (GBM) is the most common malignant brain tumor in adults and has a median survival of 13.4 months. Convection enhanced delivery (CED) has shown promise for the treatment of GBM by allowing intratumoral delivery of therapeutics, bypassing the blood brain barrier. A fiberoptic microneedle device (FMD) CED catheter enables simultaneous delivery of laser energy and therapeutic. The laser allows for heating of tissue in the region of infusion, called thermotherapy. Thermotherapy offers the advantages of increasing the volume of distribution (Vd) of the infusate, as well as facilitating intracellular penetration of the therapeutic. We hypothesize that heating of brain tissue will increase infusate Vd in ex vivo CED brain infusions. Methods: Formalin fixed mouse brains were infused by FMD-CED with Evans blue for 1 hour at 0.1 μl/min, at 22°C, 37°C and 42°C (n=4 brains/group). The Vd was determined and compared using one-way ANOVA. Results: FMD-CED performed at 42°C resulted in significantly higher mean Vd (4.90+2.2mm3; p =0.03) than those at 22°C (1.49+0.4 mm3), although no differences in Vd were observed between the other temperature groups. 42°C brains demonstrated interstitial and intracellular distribution, while rare intracellular distribution was noted in the other groups. Discussion: The Vd of FMD-CED infusions is facilitated by sub-lethal thermotherapy. This study indicates that thermotherapeutic enhancement of infusate Vd does not occur exclusively via vascular mechanisms. Thermotherapy facilitates advective-diffusion by decreasing interstitial fluid pressure and increasing transcellular fluid transport. These results were validated in a companion in vivo FMD-CED study in the rodent brain. / Master of Science

Glioblastoma is a deadly brain cancer with discouraging standard of care. New methods like convection enhanced delivery and chimeric antigen receptor T cells (CAR-T) are promising treatments that can be translated to glioblastoma. In this study, CAR-T cell flow through a hydrogel was explored in the context of in-vitro convection enhanced delivery. A culture method to create large spheroids mimicking tumors from preexisting glioblastoma stem cell lines was fabricated, a convection enhanced delivery system for in-vitro testing was designed, and characterization of the CAR-T cells using the in-vitro system took place. The spheroid culture method was successfully optimized to produce spheroids large enough to act as a sufficient tumor in little time, the in-vitro set-up successfully administered treatment, and CAR-T cells were found to increase their velocities through a medium as their injection velocity increased. It was discovered that the density of the spheroid plays a crucial role in treatment delivery, often times driving how treatment will move through the spheroid. This system can be used in the future studies to test the killing potential of CAR-T cells to a tumor in-vitro. / Master of Science / Glioblastoma is a deadly brain cancer with current treatments that are discouraging at best. New methods must be utilized to aid in patient recovery. Chimeric antigen receptor T-Cells (CAR-T) are a promising treatment that can be used in glioblastoma. In this study, CAR-T cell behavior is defined in the context of in-vitro convection enhanced delivery. A large spheroid, or sphere of cells, mimicking a tumor was created, a convection enhanced delivery system set-up for in-vitro testing was designed, and characterization of CAR-T cell behavior using the in-vitro system took place. The spheroids were successfully cultured to act as a sufficient tumor, the in-vitro set-up successfully administered treatment, and CAR-T cells were found to increase their velocities in a gel as their injection velocity increases. It was discovered that the density of the spheroid plays a crucial role in treatment delivery, often times driving how treatment will move through the spheroid. This system can be used in the future studies to test the killing potential of CAR-T cells to a tumor in-vitro.

Glioblastoma (GBM) is the most common adult malignant glioma (MG) variant, and the median survival of persons with GBM is about 2 years, even with aggressive treatments. Dogs and humans are the only species in which brain tumors commonly develop spontaneously, with an estimated post-mortem frequency of primary brain tumors approximating 2% in both species. Gliomas represent about 35% of all canine primary brain tumors, with high-grade oligodendroglioma and astrocytoma phenotypes accounting for about 70% of all canine gliomas. Canine gliomas are also treated using surgical, radiotherapeutic, and chemotherapeutic regimens similar to those used in humans. The efficacy of these therapies in dogs with MG is also poor, with median survival times ranging from 3-8 months, which closely mirrors the dismal prognosis associated with human GBM. Thus, treatment of MG represents a current and critically unmet need in both human and veterinary medicine. In this work, we investigate minimally invasive methods to access the brain for the purposes of ultimately improving the diagnosis and treatment of malignant brain tumors. Chapter 1 reviews the current clinical challenges associated with the treatment of GBM, highlights the value of using the spontaneous canine glioma model in translational brain tumor studies, and introduces High-Frequency Irreversible Electroporation (H-FIRE) and Convection Enhanced Delivery (CED), which are two novel treatment platforms for GBM being developed in our lab. In Chapter 2, we demonstrate that definitive diagnosis of brain tumors, a critical first step in patient management, can be safely and accurately performed in dogs with naturally occurring brain tumors using a stereotactic brain biopsy procedure. Chapter 3 evaluates the in vivo safety and biocompatibility of fiberoptic microneedle devices, a major technical component of our convection-enhanced thermotherapy catheter system (CETCS), chronically implanted in the rodent brain. The CETCS is a novel technology being developed and used in our laboratory to improve the delivery of drugs to brain tumors using CED. This study provides regulatory data fundamental to the commercialization of the CETCS device for brain tumor treatment by illustrating that the device did not cause clinically significant neurological complications and resulted in mild pathologic changes in brain tissue, similar to other types of devices designed and approved for use in the brain. In Chapters 4 and 5 we explore possible bystander effects of H-FIRE on glutamate metabolism in the brain. H-FIRE has been shown to be able to both ablate brain tumors as well as disrupt the blood-brain barrier (BBB). As these therapeutic effects of H-FIRE are dependent on applying electrical fields to the tissue that either reversibly permeabilize the cell membrane, allowing treated cells to survive, or permanently disrupt the structure of the cell membrane, causing cell death, we hypothesized that altering the membrane permeability with HFIRE would increase the extracellular glutamate concentrations and contribute to excitotoxic brain tissue damage. Chapters 4 used in vitro brain cell culture systems and in vivo experiments in normal and glioma-bearing rat brains to determine if glutamate release in the brain occurs as a bystander effect following H-FIRE treatment, identify concentrations of glutamate necessary to induce death of cells or BBB disruption, and characterize glutamatergic gene expression in response to H-FIRE treatment. Chapter 5 describes the use of magnetic resonance spectroscopic and spatial transcriptomic methods to further quantify the in vivo effects of H-FIRE treatment on glutamate release and metabolism in dogs with spontaneous brain tumors. The in vitro results indicated that the magnitude of glutamate release following H-FIRE is insufficient to induce cytotoxicity in normal or neoplastic brain cell lines, and also did not increase the permeability of the BBB. In our in vivo model systems, we documented significant, transient post-H-FIRE increases in glutamate to concentrations previously associated with excitotoxicty, with upregulation of the expression of genes involved with ionotropic and metabotropic glutamatergic receptor signaling. A contemporaneous upregulation of genes associated with glutamate uptake and recycling were also noted, indicating an adaptive, protective response to the glutamate release. Our work summarily demonstrates that the diagnosis and potential treatment of malignant brain tumors can be achieved through the use of minimally invasive techniques that provide local access to brain tissue. While complications will always be possible anytime the brain is manipulated surgically, and further investigations are required to characterize the spectrum and mechanisms of adverse events that can occur following CETCS CED and H-FIRE treatment, our results support the continued development of these novel therapeutic platforms for the treatment of GBM. / Doctor of Philosophy / Glioblastoma (GBM) is the most common adult malignant glioma (MG) variant, and the median survival of persons with GBM is about 2 years, even with aggressive treatments. Dogs and humans are the only species in which brain tumors commonly develop spontaneously, with an estimated post-mortem frequency of primary brain tumors approximating 2% in both species. Gliomas represent about 35% of all canine primary brain tumors, with high-grade oligodendroglioma and astrocytoma phenotypes accounting for about 70% of all canine gliomas. Canine gliomas are also treated using surgical, radiotherapeutic, and chemotherapeutic regimens similar to those used in humans. The efficacy of these therapies in dogs with MG is also poor, with median survival times ranging from 3-8 months, which closely mirrors the dismal prognosis associated with human GBM. Thus, treatment of MG represents a current and critically unmet need in both human and veterinary medicine. In this work, we investigate minimally invasive methods to access the brain for the purposes of ultimately improving the diagnosis and treatment of malignant brain tumors. Chapter 1 reviews the current clinical challenges associated with the treatment of GBM, highlights the value of using the spontaneous canine glioma model in translational brain tumor studies, and introduces High-Frequency Irreversible Electroporation (H-FIRE) and Convection Enhanced Delivery (CED), which are two novel treatment platforms for GBM being developed in our lab. In Chapter 2, we demonstrate that definitive diagnosis of brain tumors, a critical first step in patient management, can be safely and accurately performed in dogs with naturally occurring brain tumors using a stereotactic brain biopsy procedure. Chapter 3 evaluates the in vivo safety and biocompatibility of fiberoptic microneedle devices, a major technical component of our convection-enhanced thermotherapy catheter system (CETCS), chronically implanted in the rodent brain. The CETCS is a novel technology being developed and used in our laboratory to improve the delivery of drugs to brain tumors using CED. This study provides regulatory data fundamental to the commercialization of the CETCS device for brain tumor treatment by illustrating that the device did not cause clinically significant neurological complications and resulted in mild pathologic changes in brain tissue, similar to other types of devices designed and approved for use in the brain. In Chapters 4 and 5 we explore possible bystander effects of H-FIRE on glutamate metabolism in the brain. H-FIRE has been shown to be able to both ablate brain tumors as well as disrupt the blood-brain barrier (BBB). As these therapeutic effects of H-FIRE are dependent on applying electrical fields to the tissue that either reversibly permeabilize the cell membrane, allowing treated cells to survive, or permanently disrupt the structure of the cell membrane, causing cell death, we hypothesized that altering the membrane permeability with HFIRE would increase the extracellular glutamate concentrations and contribute to excitotoxic brain tissue damage. Chapters 4 used in vitro brain cell culture systems and in vivo experiments in normal and glioma-bearing rat brains to determine if glutamate release in the brain occurs as a bystander effect following H-FIRE treatment, identify concentrations of glutamate necessary to induce death of cells or BBB disruption, and characterize glutamatergic gene expression in response to H-FIRE treatment. Chapter 5 describes the use of magnetic resonance spectroscopic and spatial transcriptomic methods to further quantify the in vivo effects of H-FIRE treatment on glutamate release and metabolism in dogs with spontaneous brain tumors. The in vitro results indicated that the magnitude of glutamate release following H-FIRE is insufficient to induce cytotoxicity in normal or neoplastic brain cell lines, and also did not increase the permeability of the BBB. In our in vivo model systems, we documented significant, transient post-H-FIRE increases in glutamate to concentrations previously associated with excitotoxicty, with upregulation of the expression of genes involved with ionotropic and metabotropic glutamatergic receptor signaling. A contemporaneous upregulation of genes associated with glutamate uptake and recycling were also noted, indicating an adaptive, protective response to the glutamate release. Our work summarily demonstrates that the diagnosis and potential treatment of malignant brain tumors can be achieved through the use of minimally invasive techniques that provide local access to brain tissue. While complications will always be possible anytime the brain is manipulated surgically, and further investigations are required to characterize the spectrum and mechanisms of adverse events that can occur following CETCS CED and H-FIRE treatment, our results support the continued development of these novel therapeutic platforms for the treatment of GBM.

Targeted drug delivery to the brain is difficult to achieve using conventional techniques, largely due to the blood-brain barrier’s (BBB) impediment to drug diffusion into the brain parenchyma. In response, development of convection-enhanced delivery (CED) offers the ability to circumvent the BBB and target specific areas of the brain. Predictability of infusate movement in pathological brain states during CED will maximize the effectiveness of this treatment, and therefore modeling of infusate movement must be characterized. Previous work from our lab effectively modeled CED in rats using the middle carotid artery occlusion model of cytotoxic edema. However, previous models examined for vasogenic edema study did not show pure vasogenic edema. The purpose of this study was to develop a model of pure vasogenic edema in the rat brain. In this study, we show that stereotactic 9 µL infusion of 1.0 mM DCA over 45 minutes into the rat corpus callosum reproducibly creates pure vasogenic edema, as observed in the peritumoral white matter surrounding gliomas.

Ce travail de thèse a pour objectif le traitement local des gliomes malins via l'administration de nanocapsules lipidiques (LNC) par convection enhanced delivery (CED). Deux types de complexes métalliques lipophiles aux propriétés thérapeutiques ont été encapsulés au sein des LNC. Le premier est un complexe radioactif de Rhénium-188 et le second, un agent anticancéreux dérivé du tamoxifène et du ferrocène, le ferrociphénol (Fc-diOH). Les LNC de 188Re permettent une rétention de l'émetteur β- au niveau local et une éradication complète de la tumeur est possible pour une dose de 8Gy puisque 33% des animaux sont de longs survivants. Cette dose optimisée s'est révélée être une dose efficace, intermédiaire entre des doses toxiques (10-12 Gy) ou inefficaces (3-4 Gy). Les LNC-Fc-diOH présentent des taux d'encapsulation élevés, et sont quantitativement internalisées dans les cellules 9L. De plus, l'activité du ferrociphénol est conservée après encapsulation et se révèle très efficace sur des cellules de gliome 9L (IC50=0.6μM). En revanche, l'activité est très réduite sur les astrocytes, cellules au potentiel de division quasiment nul. L'action intratumorale des LNC-Fc-diOH dans un modèle de gliome sous-cutané entraîne une réduction significative des masses et volumes tumoraux. De plus, l'association entre le ferrociphénol et les photons X est une association synergique conférant à Fc-diOH des propriétés de molécule radio-sensibilisante. La médiane de survie du groupe traité par une CED de LNC-Fc-diOH suivie d'une radiothérapie externe de 18Gy (3x6Gy) augmente de 48% par rapport au groupe contrôle avec la présence de 17% de longs survivants.

Les gliomes de haut-grades sont des les tumeurs primitives les plus fréquentes du système nerveux central. Le traitement de cette pathologie associe chirurgie, radiothérapie et chimiothérapie. Les principales faiblesses de ces traitements sont le caractère infiltrant de la tumeur au sein d’un parenchyme hautement fonctionnel, l’existence de la barrière hémato-encéphalique limitant le passage trans-vasculaire de la chimiothérapie et la radiorésistance naturelle des cellules gliomateuses.Parmi les stratégies proposées pour outre-passer cette barrière hémato-encéphalique, une injection directe au sein du parenchyme a été évoquée. Afin d’optimiser cette délivrance le concept d’infusion convective a été développé, il s’agit d’une injection intra-parenchymateuse à un débit lent et contrôlé.Le bevacizumab est un anticorps dirigé contre le VEGF-A, un des principaux facteurs angiogéniques. Le but de ce traitement est de lutter contre l’ angiogénèse et de freiner la croissance tumorale.Dans un premier temps, la pharmacocinétique d’une injection intracérébrale de bevacizumab a été étudiée en comparaison avec une administration systémique plus classique. Les résultats permettent de mettre en évidence une concentration locale équivalente avec des concentrations systémiques beaucoup plus faibles avec une injection intra-tumorale. Un point important de cette étude est que la concentration dans l’hémisphère controlatéral à l’injection est aussi importante que lors d’une injection systémique.Puis l’efficacité d’une injection intratumorale de bevacizumab a été comparée à un traitement systémique sur un modèle de gliome chez la souris. L’efficacité du traitement est claire sur la survie de l’animal avec un avantage pour une injection intratumorale par rapport à une injection systémique. D’un point de vue microscopique cet avantage de survie peut être corrélé à une angiogénèse et une prolifération tumorale moins importante an cas d’injection directe au sein de la tumeur.Contrairement aux études pré-cliniques chez les rongeurs, les principaux essais cliniques n’ont pas permis de mettre en évidence un avantage d’une injection intra-tumorale directe. Principalement du à une mauvaise délivrance liée à des fuites et des reflux. Une des limites du modèle petit animal est l’absence de sillon cortical, vecteur de fuite. Le développement d’un modèle de gliome anatomiquement pertinent permettrait de simuler au mieux ces fuites et simultanément la mise au point de technologies de délivrance implantable à l’échelle humaine. Nous avons donc développé le premier modèle de gliome chez le porc. L’immunotolérance a été induite par un traitement par ciclosporine, des cellules de gliomes humains U87 et G6 ont été implantés, permettant se développer des tumeurs.Afin de dépister une mauvaise délivrance et anticiper les fuites ou les reflux, nous avons étudié les profils de pression le long de la ligne d’injection corrélés à l’existence de fuites ou de reflux. Nous avons pu identifier un profil pressionnel typique d’une injection de qualité. Les injections ne répondant pas à ces critères ont systématiquement conduits à des fuites ou reflux.L’étape suivante a été l’injection au sein d’une tumeur chez des porcs grâce à un système innovant implanté. Cette injection a été possible sans complication infectieuse avec une bonne tolérance locale et neurologique.La dernière étape de ce travail est l’étude anatomique de la diffusion d’un colorant injecté par une technique d’infusion convective. Cette étude s’intéressait notamment à la diffusion depuis la corona-radiata vers les différentes voies de substance blanche. La diffusion est anisotrope le long des fibres de substance blanche cependant la diffusion suit des voies différentes en fonction de la position du cathéter par rapport à elle. L’injection semble ouvrir des voies d’impédances rhéologiques faibles préférentielles nécessitant une adaptation anatomique aux voies qui seront la cible du traitement. / High grade gliomas are the most frequent primitive central nervous system tumor. The standard treatment is an association of surgery, radiotherapy and chemotherapy. The mains issues with these treatments are the infiltrative properties of the tumour in a highly functional parenchyma, the blood-brain barrier limiting the transvascular transport of chemotherapy and the inherent radioresistance of glioma cells.Upon different strategy to overpass the blood-brain barrier, a direct injection in the brain was advocated. In order to maximize this delivery, the concept of convection enhanced delivery was developed; it consists in a direct injection in the parenchyma with a low flow-rate.Bevacizumab is an anti-VEGF A antibody, VEGF is one of the most important angiogenic factors. The goal of this treatment is to inhibit the angiogenesis and slow down the tumor growth.We propose to study the use of this antibody in a direct intra-cerebral infusion.First, we focalize on the pharmacokinetic properties of an intratumoral injection by convection –enhanced delivery compared to a systemic administration. This shows an equivalent intratumoral concentration with systemic concentrations significantly lower with the intra-tumoral injection. An important result is the similar concentration in the controlateral hemisphere with the two routes of infusion. Convection-enhanced delivery is suitable to carry far from the infusion site high molecular weight proteins. An intra-tumoral bevacizumab may theoretically provide similar efficiency with less systemic side-effect.Then, the efficiency of an intra-tumoral infusion of bevacizumab is compared to a systemic injection on a mouse glioma model. In terms of survival the intra-tumoral treatment is significantly more efficient with an important decrease of angiogenesis and tumoral proliferation.If convection-enhanced delivery rodent study were promising, clinical trials failed to show any efficiency of intra-tumoral injection mainly due to inadequate delivery secondary to backflows and leakages. One of the limits of the rodent model is the absence of cortical sulci, main leakage provider. The development of a model anatomically relevant could simulate real conditions of injection and develop implantable device of injection in realistic conditions. We have developed the first induced model of glioma in a large animal. We choose the pig for the similarity of its brain anatomy and its size. The animals have been treated with ciclosporin to induce an immunosuppression, human glioma cells have been implanted, leading to the development of brain tumor.We have studied the pressure on the infusion line and correlate it to backflow and leakage. We have identified a pattern of pressure for successful infusion. Different pressure pattern have systematically led to backflow or leakage. These pressures criteria could permit to us an early detection of inadequate infusion to replace the catheter and avoid the failure of precedent clinical trials.Next step have been the intra-tumoral injection via an implanted device on pig glioma model. No infectious complication has been related with a good local and neurologic tolerance. The injections have led to a relevant diffusion through the tumor with a rapid flow to the periphery due to the interstitial pressure gradient between the tumor and the periphery.Last step of this work have been the anatomical study of a dye distribution by convection-enhanced delivery in a human encephalon. Indeed if pig brain is similar to human brain, human white matter structure is unique. This work is focalized on the diffusion from the corona-radiata to the main white matter tracts. The distribution is anisotropic following white matter but the diffusion is different depending on the position of the catheter. The infusion seems to open low rheological impedance paths the position of the catheter have to be adapted to the white matter tract to target.

Glioblastoma multiforme (Glioma) is an extremely aggressive tumor that arises from intrinsic glial cells in the central nervous system (CNS)5. It is the most common primary brain tumor in humans and has a typical survival time of 15-16 months5. Current treatments for gliomas include surgery, radiation, and treatment with temozolomide (TMZ). While these treatments tend to add 2-3 months to a patient’s survival, none have been capable of altering the course of the disease1. One of the shortcomings of novel therapeutics for glioma, is the inability to evaluate in real time how therapeutics are affecting the patient. There is also the problem of the blood brain barrier (BBB), which can be overcome by administering drugs through an intracranial catheter (delivered via CED). The primary obstacle that’s been observed in intracranial drug delivery is the inadequacy of the delivery. This inadequacy is an inability of the drug to diffuse homogenously throughout the tumor. CED also creates a possibility for toxicity due to highly concentrated volumes of drugs delivered; we believe the novel delivery method we are trying to develop, will make this possibility null. The purpose of this study was initially to demonstrate the problem of delivering drugs via diffusion while simultaneously collecting biomarkers to interpret efficacy of the drugs, due to differing molecular weights. The other objective of this study was to demonstrate that it is possible to manipulate both the direction of bulk flow and the rate of diffusion of drugs delivered through a catheter using a gel phantom as a representative of brain tissue. What we found is that by utilizing a two-catheter method with convection and retro-convection enhanced delivery, we could in fact manipulate these parameters and achieve a more even distribution of drug (represented by fluorophores in our experiments). Using these two catheter methods, we will also be able to collect fluids from the tumor to monitor the effect of any treatment in real time.

*1-Dimensional Advective-Diffusion Model in Porous Media Infusion of therapeutic agents into tissue is makes use of two mass transport modes: advective transport, and molecular diffusion. Bulk infusion into a 0.6% wt agarose phantom was modeled as an infinite, homogenous, and isotropic porous medium saturated with the same solvent used in the infused dye tracer. The source is assumed to be spherical and isotropic with constant flow rate and concentration. The Peclet numberdecreases with power function Pe = 15762t0.337 due to the decrease in mean dye-front pore velocity as V goes to Vfinal. Diffusive mass transport does not become significant during any relevent time period. *Arborizing Fiberoptic Microneedle Catheter We have developed an arborizing catheter that allows multiple slender fused-silica CED cannulae to be deployed within a target volume of the brain via a single needle tract, and tested it in a widely accepted tissue phantom. The arborizing catheter was constructed by bonding and encapsulating seven slender PEEK tubes in a radially symmetric bundle with a progressive helical angle along the length, then grinding a conicle tip where the helical angle is greatest. The catheter was tested by casting 0.6% wt agarose around the device with all needles deployed to a tip-to-tip distance of 4 mm. Phantom temperature was maintained at 26 ± 2°C. 5% wt Indigo Carmine dye was infused at a rate of 0.3 uL/min/needle for 4 hours. N=4 infusions showed a Vd/Vi of 139.774, with a standard deviation of 45.01. This is an order of magnitude greater than single-needle infusions under similar conditions [45]. The arborizer showed the additional benefit of arresting reflux propagating up the lengths of individual needles, which has historically been a weakness of single-needle CED catheter designs. *In Vivo Co-Delivery of Single Walled Carbon Nano-horns and Laser Light to Treat Human Transitional Cell Carcinoma of the Urinary Bladder in a Rodent Model Using a rodent model we explored a treatment method for Transitional Cell Carcinoma (TCC) in the urinary bladder in which Single Walled Carbon Nanohorn (SWNH) solution and 1064 nm laser light are delivered into tumorous tissue via a co-delivery Fiberoptic Microneedle Device (FMD). Preliminary treatment parameters were determined by injecting SWNH solutions with concentrations of 0 mg/mL, 0.17 mg/mL, or 0.255 mg/mL into ex vivo porcine skin and irradiating each for three minutes at laser powers of 500 mW, or 1000 mW. The combination with the greatest temperature increase without burning the tissue, 0.17 mg/mL at 1000 mW, was selected for the in vivo treatment. TCC tumors were induced in a rodent model by injecting a solution of 106 AY27 urothelial carcinoma cells into the lateral aspect of the left hind leg of young, female F344 rats. When tumors reached 5-10 mm3, rats were anesthitized and treated. SWNH solution was injected directly into the tumor and irradiated until the target temperature of 60degC was achieved. The rats were then recovered from anestesia and monitored for 7-14 days, at which point they were humanely sacrificed, and the tumors prepared for histological examination. Histological assessment of areas of FMD treatment correlated well with gross morphological appearance. Foci of tumor necrosis showed sharp (1-2 mm) delineation from areas of viable tumor (not treated) and normal tissue. We believe we have demonstrated the feasibility of using the FMD for treatment of urothelial carcinoma using an animal model of this disease, and are encouraged to continue development of this treatment and testing in larger animal models. / Master of Science

Alzheimer's disease (AD), first characterized in the early 20th century, is a common form of dementia which can occur as a result of genetic mutations in the genes encoding presenilin 1, presenilin 2, or amyloid precursor protein (APP). These genetic alterations can accelerate the pathological characteristics of AD, including the formation of extracellular neuritic plaques composed of amyloid beta peptides and the formation of intracellular neurofibrillary tangles consisting of hyperphosphorylated tau protein. Ultimately, AD results in gross neuron loss in the brain which is evidenced clinically as a progressive decline in mental capacity. A strong body of scientific evidence has previously demonstrated that the driving factor in the pathogenesis of AD is potentially the accumulation of Aß peptides in the brain. Thus, reduction of Aß deposition is a major therapeutic strategy in the treatment of AD. Recently it has been suggested that Aß accumulation in the brain is modulated, not only by Aß production, but also by its degradation. Several important studies have demonstrated that Aß degradation is modulated by several endogenous zinc metalloproteases shown to have amyloid degrading capabilities. These endogenous proteases include neprilysin (NEP), endothelin converting enzyme (ECE), insulin degrading enzyme (IDE) and matrix metalloprotease 9 (MMP9). In this investigation we study the effects of upregulating expression of several of these proteases through administration of recombinant adeno-associated viral vector (rAAV) containing both endogenous and synthetic genes for ECE and NEP on amyloid deposition in amyloid precursor protein (APP) plus presenilin-1 (PS1) transgenic mice. rAAV administration directly into the brain resulted in increased expression of ECE and NEP and a substantial decrease in amyloid pathology. We were able to significantly increase the area of viral distribution by using novel delivery methods resulting in increased gene expression and distribution. These data support great potential of gene therapy as a method of treatment for neurological diseases. Optimization of gene transfer methods aimed at a particular cell type and brain region in the CNS can be accomplished using AAV serotype specificity and novel delivery techniques leading to successful gene transduction thus providing a promising therapeutic avenue through which to treat AD.

Abstract : Glioblastoma is the most common and aggressive brain cancer in adults. The current standard-of-care treatment includes surgical resection, radiation therapy with concomitant and adjuvant temozolomide (TMZ) chemotherapy. However, the addition of TMZ to radiation therapy only increased the median survival time (MeST) by 2.5 months. This limited improvement is partially attributable to the low accumulation of chemotherapeutic drugs in the brain tumor due to the blood-brain barrier (BBB). Thus, new delivery methods such as intra-arterial, BBB disruption and convection-enhanced delivery (CED) have been proposed to overcome this limitation. Besides, timing tumor irradiation to coincide with the maximal concentration of platinum-DNA adducts could result in improved tumor control. In this study, CED of cisplatin and oxaliplatin, their respective liposomal formulations Lipoplatin™, Lipoxal™, and carboplatin with or without 15 Gy of radiation therapy has been carried out in F98 glioma bearing Fischer rats to assess their toxicity and MeST. The amount of platinum-DNA adducts in the tumor at 4 h and 24 h after CED was measured and irradiation was administered at these two different time periods to test the concomitant effect. In addition, four liposomal carboplatin formulations with different chemo-physical properties were prepared and their toxicity and MeST were also evaluated in this animal model. Among the tested platinum drugs, carboplatin and Lipoxal™ demonstrated a highest maximum-tolerated dose of 25 µg and 30 µg respectively. CED of carboplatin showed the longest MeST of 38.5 days, and increased to 54.0 days with the addition of 15 Gy radiation therapy. However, radiation at 4 h after CED of either oxaliplatin or carboplatin did not show any survival improvement when compared to radiation at 24 h, although the quantity of platinum-DNA adducts at 4 h was higher than at 24 h after CED. In the four liposomal carboplatin formulations, anionic pegylated liposomal carboplatin showed the longest MeST of 49.5 days, due to its longer tumoral retention time and probably larger distribution volume in the brain. / Résumé : Le glioblastome est le cancer primaire du cerveau le plus courant et agressif chez l’adulte. Le traitement standard comprend la résection chirurgicale, la radiothérapie et la chimiothérapie concomitante et adjuvante avec le témozolomide(TMZ). L'addition de TMZ combinée la radiothérapie a augmenté la survie médiane (MeST) de 2,5 mois. Cette faible amélioration est partiellement due à l'accumulation limitée de médicaments chimiothérapeutiques dans la tumeur cérébrale à cause de la barrière hémato-encéphalique (BBB). Ainsi, de nouvelles méthodes comme l’injection intraartérielle, la rupture osmotique de la barrière hémato-encéphalique, la livraison augmentée par convection (CED) ont été suggérées pour surmonter ce problème. En plus, l’optimisation de l’irradiation de la tumeur lorsque le maximum d’adduits platine-ADN est atteint pourrait aboutir à un meilleur contrôle de la tumeur. Dans cette étude, nous avons injecté par CED le cisplatine, l’oxaliplatine, avec leur formulation liposomale Lipoplatin™, Lipoxal™ ainsi que le carboplatine avec ou sans radiation de 15 Gy. La toxicité et le temps de MeST ont été mesurés chez des rats Fischer porteurs du gliome. La quantité d'adduits platine-ADN dans la tumeur a été mesurée 4 h et 24 h après CED. L’irradiation de la tumeur a été effectuée à ces deux temps pour tester l'effet concomitant. En plus, quatre formulations liposomales de carboplatine avec différentes propriétés chimiophysiques ont été préparées et leur toxicité et MeST combiné à la radiation ont également été évalués. Parmi les drogues de platine testées, le carboplatine et Lipoxal™ ont démontré respectivement la dose maximale tolérée la plus élevée, soit 25 µg et 30 µg. La MeST du carboplatine était la plus longue avec 38,5 jours qui a augmenté à 54,0 jours avec l’addition de 15 Gy de radiothérapie. Toutefois, l’irradiation à 4 h après CED effectuée avec l'oxaliplatine et le carboplatine n'a pas amélioré la MeST comparé à l’irradiation à 24 h, bien que la quantité d'adduits platine-ADN à 4 h était supérieure à celle mesurée à 24 h après CED. Pour les quatre formulations liposomales de carboplatine, celle pégylée négatif a démontré la plus longue MeST, soit 49,5 jours.