Rozprawy doktorskie na temat „Preclinical tumor models”

One of the recent advancements in oncological research has been the recognition of the tumor microenvironment (TME) as a relevant participant during all stages of the evolution of a neoplastic process. Indeed, over the past decades, tumors have been considered through a changing perspective: no longer as a growth of homogeneous neoplastic cells, but as an actual organ composed of different cell populations and structures: the parenchyma being the neoplastic population and the stroma, including the vascular network and infiltrating cells. The tumor microenvironment has a dual role in tumor biology, both promoting and antagonizing tumor development, growth, and local or distant invasiveness. According to its leading role in influencing tumor biology each component of the TME could be considered as a potential pharmacological target to be enhanced or antagonized, in order to influence tumor behavior. Accordingly, the study of the TME could provide new insights in the tumor biology and offers numerous potential targets for the development of novel therapeutic strategies. In this context, morphological techniques represent useful tools for the investigation of the TME, allowing the evaluation of the spatial distribution of the different elements, and provide useful complementary information to clinical and other data obtained in experimental in vivo studies. In this thesis, the three main classes of the TME components -tumor-associated vasculature, immune-inflammatory cells and tumor stroma- are illustrated in three different chapters and relevant experimental studies described. However, it should be considered that the various aspects of TME are not separate entities but are all involved in a dynamic system with complex structural and functional interactions. Chapter 1 – Tumor-associated vasculature Tumor angiogenesis has been identified as a hallmark of cancer, due to its central role in supporting tumoral growth, providing nutrient supply, removing catabolites and enabling tumoral metastatic dissemination. Most of the solid tumors are characterized by an “angiogenetic switch” in which an imbalance between pro- and anti-angiogenic factors sustains a dysregulated angiogenetic process, leading to the formation of an altered vascular network composed of structurally and functionally abnormal blood vessels. Drugs targeting tumor vasculature has been extensively studied as a mean to interfere with tumoral growth as well as to promote the delivery and/or effect of co-administered compounds to the tumor. In the first study of this chapter, we demonstrated the therapeutic efficacy and the antiangiogenic effect of a novel compound developed by binding sunitinib (a well-known antiangiogenic drug) to a selective binder of αVβ3 integrin thus promoting its delivery to the target site (tumors expressing αVβ3 integrin). The other studies of this chapter investigated the relation between tumor vasculature and tumor hypoxia. In particular, this relation was investigated to uncover the potential mechanism underlying the synergistic effect of the administration of an antiangiogenic compound (cediranib) with a poly-ADP ribose polymerase (PARP) inhibitor (olaparib) in a panel of patient-derived xenografts of ovarian carcinoma. Chapter 2 – Tumor immune microenvironment In most cancers, both innate and acquired immunity have a driving role during all stages of tumor development and progression. Depending on the cell population and/or molecular stimuli received, they can act in a dual way, antagonizing or promoting tumor growth. Three selected studies were described in chapter 2 and investigated: 1. The role of NK cells in hindering metastasis engraftment in a metastatic model of synovial sarcoma. After the combined administration of an heparanase-inhibitor with a tyrosine kinase inhibitor a significant reduction of lung metastases was observed and immunohistochemical analyses demonstrated the role of NK cells in this phenomenon. 2. The macrophage polarization status in a panel of xenotransplanted thyroid carcinoma tumors. The mononuclear-phagocyte populations infiltrating the tumors were evaluated by immunohistochemistry. 3. The role of inflammation in the development of colorectal cancer was evaluated in mice (wild type and EMILIN1-mutant), undergoing administration of AOM-SS (chemical carcinogenesis model). EMILIN1 mutant mice developed more numerous and more severe tumoral lesions compared to wild type, as well as increased inflammatory infiltrate was observed, unveiling a potential contribution of Emilin 1 in the pathogenesis of colorectal adenocarcinoma. Chapter 3 – Tumor stroma Tumor stroma represents not only the scaffold in which tumors growth, but also an intricate network of molecules and signals influencing tumor biology. The first study of this chapter investigated stroma-derived circulating molecules as a potential tool for the early diagnosis of pancreatic ductal adenocarcinoma (PDAC). Selected molecules (MMP-7, TIMP-1 and Throbospondin-2) were tested in KC genetically engineered mice (modeling the early stages of PDAC development) and patient-derived xenografts (modeling tumor progression), by serum ELISA and by immunohistochemistry. The second study evaluated the potential improvement in the biodistribution of chemotherapeutic drugs derived from the combined treatment with hyaluronidase. Tumor-bearing mice (ovarian carcinoma and pancreatic carcinoma models) were treated with chemotherapy alone (paclitaxel) or combined with hyaluronidase. Hyaluronidase treatment reduced the amount of stromal hyaluronic acid (as demonstrated by Alcian blue stain) and improved intratumor distribution of paclitaxel (as analyzed by mass spectrometry).

Tumor acidosis is an important biomarker in cancer. We have developed a noninvasive imaging method, termed acidosis Chemical Exchange Saturation Transfer (acidoCEST) MRI to measure extracellular pH (pHe) in the tumor microenvironment. Chapter 1 introduces the importance of measuring tumor acidosis and presents various imaging modalities and their shortcoming to measure pHe. Chapter 2 describes the optimization of acidoCEST MRI for in vivo pHe measurement. The acidoCEST MRI protocol consists of a CEST-FISP acquisition and Lorentzian line shape fittings. We determined the optimal saturation time, saturation power and bandwidth, 5 sec, 2.8 µT and 90 Hz respectively. We also tried various routes of administration to increase contrast agent uptake in the tumor. We decided upon 200 µL bolus followed by 150 µL/hr infusion. The optimized acidoCEST MRI protocol was tested on a mammary carcinoma mouse model of MDA- MB-231. Our method can detect an increase in pHe in the bladder and tumor of the mice treated with bicarbonate. We used this optimized acidoCEST MRI method to measure pHe in lymphoma tumor model of Raji, Ramos and Granta 519 as described in Chapter 3. Pixel-wise pHe maps showed tumor heterogeneity. The pHe of Raji, Ramos and Granta 519 were determined to be mildly acidic with no significant difference. Chapter 4 describes the evolution of pixel-wise analysis in more detail. Besides the pHe map and spatial heterogeneity, we were able to determine the % contrast agent uptake. We monitored these biomarkers in two different mammary carcinoma mouse models, MDA- MB-231 and MCF-7 longitudinally and made comparisons between the different tumor models: MCF-7 were more acidic, more heterogeneous and faster growing than MDA- MB-231.

Il processo di sviluppo di un nuovo farmaco oncologico e' caratterizzato da un elevatissimo numero di fallimenti, principalmente dovuti alla scarsa efficacia o eccessiva tossicita' riscontrata durante le fasi di sperimentazione clinica. Tra le possibili cause di questo fenomeno vi sono l'utilizzo di modelli animali poco rappresentativi della condizione umana e la mancanza di un paradigma di traslazione dal contesto preclinico a quello clinico sufficientemente predittivo. L'utilizzo di modelli farmacometrici, capaci di estrapolare, sintetizzare e integrare le informazioni raccolte durante la sperimentazione preclinica, puo' essere un promettente tentativo di rispondere a queste problematiche. Inserendosi nell'ambito dell'uso della modellistica matematica a supporto del processo di sviluppo di nuovi farmaci antitumorali, questa tesi si concentra sulla costruzione, implementazione ed analisi di nuovi approcci matematici per l'analisi di dati sperimentali tradizionalmente ottenuti durante le fasi di sperimentazione preclinica. Focus specifico di questo lavoro e' la modellizzazione delle interazioni tra tumore e organismo ospitante durante la somministrazione di trattamenti antitumorali resa possibile dall'utilizzo di un set di leggi di bilancio energetico fornite dalla Dynamic Energy Budget theory. L'approccio proposto, opporunatamente declinato in diversi contesti sperimetali, e' capace di tenere simultaneamente in considerazione i differenti aspetti che caratterizzano gli studi di crescita tumorale in vivo: l'effetto citotossico o citostatico della terapia antitumorale sulle cellule tumorali, l'eventuale insorgere di fenomeni di cachexia come conseguenza del trattamento o del tumore stesso infine, l'influenza che la condizione dell'organismo ha sulla crescita tumorale. Piu' nel dettaglio, e' stato sviluppato ed analizzato dal punto di vista matematico un modello di interazione tumore-organismo capace di descrivere sia l'effetto inibitorio sulla crescita tumorale di un trattamento citotossico, sia il suo effetto tossico sull'organismo ospitante. Tale modello e' stato, in particolare, adottato per analizzare dati sperimentali circa l'effetto del farmaco etoposide su ratti Wistar. E' stato inoltre sviluppato un secondo modello, sempre basato sulle interazioni energetiche di tumore e organismo, per descrivere l'effetto citostatico di un trattamento anti-angiogenico. L'approccio meccanicistico alla base del modello proposto ha permesso di tenere in considerazioni l'insorgere di fenomeni di resistenza mediata dalla condizione di ipossia tumorale in seguito a trattamenti prolungati di bevacizumab. Infine, utilizzando i modelli di inibizione di crescita tumorale precedentemente sviluppati, e' stato proposto un nuovo approccio per analizzare esperimenti di combinazione e riconoscere eventuali interazioni tra farmaci anti-angiogenici e chemioterapici.<br>The anticancer drug development process is characterized by the highest attrition rates in the clinical setting, primarily due to adverse efficacy and safety results. Preclinical animal models slightly representative of the human condition and an inadequate predictive paradigm of preclinical to clinical translation may be likely causes of this. Pharmacometric models, able to extract, synthesize and integrate preclinical information, could support the transfer of the preclinical results to the clinical setting. Within the paradigm of the Model-Informed Drug Discovery and Development, my thesis deals with the development, implementation and analysis of new mathematical modeling approaches to exploit data routinely generated in the preclinical phases of anticancer drug development process. In all the described research activities it can be recognized the importance of PK/PD modeling in better characterizing, understanding and predicting PK/PD behaviour of oncology agents. The focus of this work is a mathematical modeling of interactions between tumor and host organism during anticancer drug treatments in preclinical experiments. To this aim, a tumor-in-host modeling approach is proposed on the basis of a set of tumor-host interaction rules taken from the Dynamic Energy Budget (DEB) theory. This framework, suitably adapted to several experimental contexts, is able to integrate the different aspects characterizing the in vivo tumor growth studies: the drug cytotoxic or cytostatic activity on the tumor, the eventually onset of cachexia due to the treatment, the effect of the tumor on the host and, viceversa, the influence of the host condition on tumor dynamics. In particular, a tumor-in-host DEB-based model describing the cachexia onset and tumor growth inhibition (TGI) after the administration of cell-killing agents has been developed, mathematically analysed and, subsequently, applied on a etoposide experiment in Wistar rats. The cytostatic anticancer effect of angiogenesis inhibitors in xenograft mice has been, also, modeled within the tumor-in-host DEB-based framework. This DEB-TGI anti-angiogenic model has proved to be extremely useful to describe and understand the complexities of an hypoxia-triggered resistance to bevacizumab. Finally, starting from the previous developed TGI models, a tumor-in-host approach to analyse combination experiments and assess possible drug-drug interaction between anti-angiogenic and chemotherapeutic agents is proposed.

This thesis is a step forward in preclinical in-vivo disease modelling, designed to find new diagnostic and therapeutic options, to ultimately improve the poor outcome of patients with primary bone cancer. Combining the principles of tissue-engineering, 3D-printing and advanced gene editing techniques the preclinical animal models developed in this thesis have important clinical implications that could shape future innovative treatment plans. Particularly the translation of a humanised osteosarcoma model from a mouse into a newly engineered severely immunocompromised rat will facilitate preclinical primary bone cancer research by opening up new experimental avenues for complex surgical resection and reconstruction models.

Overexpression of huHER2 occurs in nearly 15–20% of breast cancers, and it is generally associated with poor patient survival. Existing therapies such as trastuzumab and lapatinib are currently used in the treatment of HER2-positive cancers, although issues with high recurrence and acquired resistance still remain. Elucidation of the molecular mechanisms underlying resistance is leading to the identification of therapies and strategies to manage resistance to HER2-targeted therapies. In addition to intrinsic and acquired resistance associated to HER2 oncogene, the induction of bypass pathways that reactivate growth factor-dependent signalling upon oncogene inhibition is likely pervasive across cancers and should be anticipated. Together, these findings underscore that many resistance mechanisms fall into predictable and therapeutically tractable themes, and can be effectively targeted with rationally designed combined therapies. It is, therefore, necessary to come back to dissect HER2 pathway and unravel key features contributing to its transforming capacity. The present thesis, is focused on the role played by HER2-loss variants and Delta16 isoform in mediating HER2 oncogenic activity and in conditioning the response to HER2 therapies in breast cancer. These HER2 phenotypes can drive differential drug responses of the tumor and of distant metastases. Thus, recent investigations on drug resistance and on tumor biology converged to the development of preclinical cancer models representative of cancer heterogeneity and able to mimic all possible scenarios observed in human tumors. In this project, thanks to the availability of several preclinical models representative of HER2 postive breast cancer, it was studied the contribute of HER2 and of its variants to cancer development and drug resistance. In addition, with the purpose of obtaining preclinical models that could best recapitulate human tumor heterogeneity a panel of breast cancer PDX was developed.

Mestrado em Bioquímica Clínica<br>O cancro do pulmão (CP) é um dos cancros mais diagnosticados a nível mundial e também um dos mais mortíferos. Atualmente, as terapias administradas a nível clínico para o tratamento do CP são ainda extremamente ineficazes e limitadas no que diz respeito ao aumento da taxa de sobrevivência dos pacientes oncológicos. Esta realidade demonstra a necessidade de investigar ativamente novas terapias para o tratamento desta neoplasia. No entanto a validação pré-clínica de terapias inovadoras para o CP tem-se revelado extremamente difícil devido à inexistência de plataformas que sejam adequadas para testes a nível laboratorial, uma vez que as culturas celulares in vitro bidimensionais (2D), recomendadas pelas agências regulatórias são incapazes de mimetizar as caraterísticas principais dos tumores humanos. Estas limitações têm originado uma fraca correlação entre a performance das terapias nos estudos in vitro e a obtida em ensaios clínicos controlados. Neste contexto, os modelos de tumores tridimensionais (3D) in vitro têm vindo a ser reconhecidos como uma solução para este problema, pois podem recapitular várias componentes do microambiente tumoral. Das várias plataformas 3D in vitro de CP investigadas atualmente muito poucas avaliaram o papel da inclusão de células estaminais mesenquimais (MSCs). Para colmatar esta lacuna, o trabalho de investigação desenvolvido no âmbito desta dissertação descreve a produção e otimização de novos modelos hétero-celulares 3D in vitro. Estas plataformas são compostas por células tumorais do CP (A549) e do seu estroma, nomeadamente fibroblastos da pele e células estaminais mesenquimais derivadas da medula óssea (BM-MSCs). Estes três tipos de células foram co-cultivadas em micropartículas poliméricas de policaprolactona revestidas por ácido hialurónico, com o objetivo de incluir este componente da matriz extracelular que se encontra presente no microambiente do CP. Esta abordagem permitiu formar a nível laboratorial microtecidos multicelulares 3D híbridos que melhor mimetizam a heterogeneidade celular das neoplasias pulmonares. Os resultados obtidos demonstraram que os microtumores formados através da técnica de sobreposição-líquida são reprodutíveis em termos de morfologia e tamanho, apresentaram núcleos necróticos, organização celular 3D e produziram proteínas do microambiente tumoral. Além destas caraterísticas, os dados obtidos através de microscopia de fluorescência revelaram que as BM-MSCs migram para o interior dos microtumores ao longo do tempo. A avaliação da citotoxicidade da Doxorubicina, um fármaco anti-tumoral rotineiramente utilizado a nível clínico, demonstrou que a inclusão de micropartículas aumenta a resistência das células tumorais em modelos homotípicos. Nos modelos tri-cultura heterotípicos a citotoxicidade foi comparável à obtida em microtumores sem micropartículas. Estes resultados evidenciam assim o papel importante dos fibroblastos e das BM-MSCs na resposta dos microtumores. Numa visão global, os modelos 3D formados recapitulam com mais exatidão o microambiente do cancro do pulmão e poderão servir no futuro como plataformas de teste para descobrir ou aperfeiçoar novas terapias, ou combinações de terapêuticas, para este tipo de neoplasia.<br>Lung cancer (LC) is one the most commonly diagnosed cancers worldwide, being also one of the deadliest. Currently, clinically administered therapies for treatment of LC are still extremely ineffective and limited in increasing oncologic patients survival rates. This reality evidences the necessity of actively investigating novel therapies for the treatment of LC. However, preclinical validation of novel therapies as revealed itself as an extremely arduous process, due to the lack of suitable laboratory testing platforms since the recommend in vitro bi-dimensional (2D) cell cultures are unable to fully mimic the main hallmarks of human tumors. In this context, in vitro tridimensional (3D) tumor models are being increasingly recognized as a solution due to their ability to correctly recapitulate several characteristics of the tumor microenvironment (TME). Amongst currently developed 3D in vitro platforms for the study of LC, few have included or studied the role of mesenchymal stem cells (MSCs). To provide further insights into this hypothesis, the research work developed in this thesis describes the production and optimization of novel heterotypic in vitro 3D models, comprised by non-small-cell lung cancer cells (A549) and stromal cells, namely skin fibroblasts (HFs), and bone-marrow derived mesenchymal stem cells (BM-MSCs). These three diverse cell populations were co-cultured in hyaluronic acid coated polymeric polycaprolactone microparticles (LbL-MPs) as to include this key extracellular matrix component of LC TME. This approach allowed the formation of 3D multicellular heterotypic microtissues (3D-MCTS) that better recapitulate the cellular heterogeneity of LC TME in the laboratory. The obtained findings demonstrate that these models formed via the liquid-overlay technique were reproducible in terms of morphology and size, presented necrotic core formation, 3D cellular organization, and deposited matrix proteins in a similar manner as in the TME. Besides this, fluorescence microscopy data revealed that BM-MSCs migrated overtime into the microtumors core . Performed doxorubicin in vitro cytotoxicity assays revealed that the inclusion of LbL-MPs lead to an increased resistance of homotypic A549 monoculture models against this anti-cancer drug commonly used in clinical treatments. Alongside, the cytotoxicity obtained in triculture heterotypic models was comparable to that of microtumors without LbL-MPs inclusion, showcasing the role of HFs and BM-MSCs in microtumors response to therapy. Globally, the herein bioengineered 3D models were able to recapitulate with an increased precision the TME of LC, making them suitable test platforms for development or improvement of standalone or combinatorial therapies for this type of neoplasia.

Cancer is a deadly, complex disease with 14 million new cases diagnosed every year and the endeavour to develop a cure is a global multidisciplinary effort. The complexity of cancer and the resulting vast volume of data derived from its research necessitates a robust and cutting-edge system of mathematical and statistical modelling. This thesis proposes novel mathematical models of quantification and modelling applied to heterogeneous preclinical cancers, focusing on the translation of animal studies into patients with particular emphasis on tumour stroma. The first section of this thesis (quantification) will present different techniques of extracting and quantifying data from bioanalytical assays. The overall aim will be to present and discuss potential methods of obtaining data regarding tumour volume, stromal morphology, stromal heterogeneity, and oxygen distribution. Firstly, a 3D scanning technique will be discusses. This technique aims to assess tumour volume in mice more precisely than the current favoured method (callipers) and record any cutaneous symptoms as well, with the potential to revolutionise tumour growth analysis. Secondly, a series of image processing methods will be presented which, when applied to tumour histopathology, demonstrate that tumour stromal morphology and its microenvironment play a key role in tumour physiology. Lastly, it will be demonstrated through the integration of in-vitro data from various sources that oxygen and nutrient distribution in tumours is very irregular, creating metabolic niches with distinct physiologies within a single tumour. Tumour volume, oxygen, and stroma are the three aspects central to the successful modelling of tumour drug responses over time. The second section of this thesis (modelling) will feature a mathematical oxygen-driven model - utilising 38 cell lines and 5 patient-derived animal models - that aims to demonstrate the relationship between homogeneous oxygen distribution and preclinical tumour growth. Finally, all concepts discussed will be merged into a computational tumour-stroma model. This cellular automaton (stochastic) model will demonstrate that tumour stroma plays a key role in tumour growth and has both positive (at a molecular level) and negative (at both a molecular and tissue level) effects on cancers. This thesis contains a useful set of algorithms to help visualise, quantify, and understand tissue phenomena in cancer physiology, as well as providing a series of platforms to predict tumour outcome in the preclinical setting with clinical relevance.

The microbiome has gained increasing momentum in cancer research, as it has become clear that microorganisms residing within our body are involved in mediating the cellular and tissue metabolism in health and disease. In bladder cancer research, there are different microbial communities that may mediate cancer pathobiology and response to therapy: the gut microbiome, the urinary microbiome, the urothelium-bound microbiome. These bacterial communities may mediate the processes of carcinogenesis or recurrence, modify the response to local intravesical therapies or influence the activity of systemic anticancer protocols. Based on these premises, my research project aimed to unveil the urinary and urothelium-bound microbiome in therapy-naïve bladder cancer patients, describing the differently enriched bacterial communities using a sex-based stratification. Compared to healthy controls, I found that the urine of men affected by bladder cancer were enriched in the order Opitutales and subordinate family Opitutaceae, together with the isolated class Acidobacteria-6, while in female patients I found enriched the genus Klebsiella. Notably, the bladder cancer tissue was enriched in the genus Burkholderia in both men and women, when compared to non-neoplastic, paired urothelium biopsies. Then, I also characterized the gut microbiome of bladder cancer patients undergoing neoadjuvant pembrolizumab to understand if the intestinal bacteria may influence the immune-mediated anticancer activity. In this set, I have reported that antibiotic therapy has a negative effect on immunotherapy efficacy. Second, the gut microbiome of patients not responding to neoadjuvant pembrolizumab was characterized by a higher abundance of Ruminococcus bromii, while patients who showed a response were enriched in the genus Sutterella. Lastly, I started the implementation of in vivo and in vitro systems to test the mechanistic role of the bacteria identified in human samples. This thesis work reported innovative data on the role of different microbial communities (urinary/urothelium-bound/fecal) in bladder cancer and bladder cancer therapy, and provided novel in vivo and in vitro models to validate those finding and uncover the complex microbiome-host cells crosstalk in bladder cancer patients.<br>The microbiome has gained increasing momentum in cancer research, as it has become clear that microorganisms residing within our body are involved in mediating the cellular and tissue metabolism in health and disease. In bladder cancer research, there are different microbial communities that may mediate cancer pathobiology and response to therapy: the gut microbiome, the urinary microbiome, the urothelium-bound microbiome. These bacterial communities may mediate the processes of carcinogenesis or recurrence, modify the response to local intravesical therapies or influence the activity of systemic anticancer protocols. Based on these premises, my research project aimed to unveil the urinary and urothelium-bound microbiome in therapy-naïve bladder cancer patients, describing the differently enriched bacterial communities using a sex-based stratification. Compared to healthy controls, I found that the urine of men affected by bladder cancer were enriched in the order Opitutales and subordinate family Opitutaceae, together with the isolated class Acidobacteria-6, while in female patients I found enriched the genus Klebsiella. Notably, the bladder cancer tissue was enriched in the genus Burkholderia in both men and women, when compared to non-neoplastic, paired urothelium biopsies. Then, I also characterized the gut microbiome of bladder cancer patients undergoing neoadjuvant pembrolizumab to understand if the intestinal bacteria may influence the immune-mediated anticancer activity. In this set, I have reported that antibiotic therapy has a negative effect on immunotherapy efficacy. Second, the gut microbiome of patients not responding to neoadjuvant pembrolizumab was characterized by a higher abundance of Ruminococcus bromii, while patients who showed a response were enriched in the genus Sutterella. Lastly, I started the implementation of in vivo and in vitro systems to test the mechanistic role of the bacteria identified in human samples. This thesis work reported innovative data on the role of different microbial communities (urinary/urothelium-bound/fecal) in bladder cancer and bladder cancer therapy, and provided novel in vivo and in vitro models to validate those finding and uncover the complex microbiome-host cells crosstalk in bladder cancer patients.

Medulloblastoma is the most common malignant brain tumour in children. The current treatment for medulloblastoma consists of surgery, radiation, and chemotherapy. Although these therapies have their merits, the outcome for some patients, particularly those with MYC amplified tumours, is poor, and the damaging nature of these therapies results in morbidities that significantly impact on a patient’s quality of life. To improve outcome and reduce adverse side effects, several strategies have been employed, including expanding the repertoire of accurate disease models, the development of novel therapies and the initiation of clinical trials, and an improvement in the understanding of the disease pathogenesis. The purpose of this thesis was to contribute to the repertoire of animal models, assist in identifying novel therapeutic approaches to treatment, and to advance the knowledge of the medulloblastoma immune microenvironment. We did so by aiming to (1) develop more accurate, immune-competent animal models of MYC- or NMYC-amplified medulloblastoma, (2) testing a novel treatment that combines conventional chemotherapies with a cell cycle checkpoint kinase inhibitor, and (3) investigating the effects of clinically used chemotherapies on immune cell populations in the brains of medulloblastoma-bearing mice. Currently, the limited availability of preclinical mouse models that accurately represent subgroup-specific medulloblastoma has hindered the development of therapies that succeed in the clinical setting. In addition, the distinct lack of immunologically competent models has prevented the advancement of immunotherapy drugs in treating medulloblastoma. The models of medulloblastoma developed here were designed to aid in preclinical studies, and to contribute specifically to the repertoire of immune competent mouse models for studies to clarify the role of the immune system in medulloblastoma. Here, we utilised mutated human variants of CMYC, NMYC, and P53, to transform mouse neural stem cells into tumour forming cells. Tumours were established in C57Bl/6 mice and characterised by histology and RNAseq analysis. Whilst these animal models could not be confirmed to accurately represent Group 3 or Group 4 medulloblastomas, this study demonstrated that mouse neural stem cells could be transformed using human genes and could generate brain tumours within immune competent hosts. As we were unable to conclusively define the exact nature of the tumours that were produced here, existing mouse models were used for subsequent chapters in this thesis. Prior to implementing new therapeutic agents into clinical trial, these are evaluated in a pre-clinical setting. The data presented in chapter two contributed to a larger scale pre-clinical testing pipeline that led to the identification and evaluation of multiple kinase inhibitors for the treatment of medulloblastoma. Here, I examined the combination of the cytotoxic agent gemcitabine (GEM) with the cell cycle checkpoint kinase inhibitor (LY2606368, prexasertib) as a novel approach in the treatment of Group 3 medulloblastoma. Combination GEM/LY2606368 treatment improved the survival of mice with aggressive medulloblastoma. Using immunoblotting and flow cytometry I showed that mechanistically LY2606368 enhanced GEM-induced cytotoxicity by impairing DNA damage response pathway activity, which promoted the accumulation of DNA damage leading to increased apoptosis. Together, these data formed part of the preclinical evidence that supported the establishment of the SJ-ELiOT clinical trial, targeted towards improving outcomes for patients who experience recurrent or relapsed disease following standard-of-care therapy. There is evidence to suggest that inhibiting the DNA damage response pathway can stimulate the immune system, and aid in tumour elimination. This provides strong rationale for implementing immunotherapies for patients who are predicted to have a poor response to conventional treatments. Unfortunately, to date all clinical trials investigating current popular immune-based therapies have failed in medulloblastoma, likely due to a poor understanding of the immune microenvironment in this disease. This presented an opportunity to improve our understanding of the effects of treatment on the immune system in brain. Here I characterised the immune cell populations in the brains of mice with Group 3 medulloblastoma treated with vehicle or two clinically-used chemotherapies – cyclophosphamide (CPA) and GEM. I revealed that CPA and GEM differentially alter immune cells within the brain in a manner similar to that observed outside of the central nervous system. I also demonstrate that the lack of an adaptive immune system (using mice deficient in Rag1) does not influence the anti-cancer effects of these drugs. This information provides a rationale for exploring alternative avenues when considering the use of cancer-targeting immunotherapies in combination with conventional medulloblastoma treatments. Collectively, these studies demonstrate the complicated nature of modelling high-risk medulloblastoma in the lab. I have improved upon current preclinical tools for medulloblastoma by the development of accurate immune competent models and advanced our knowledge of disease pathogenesis by elucidating the way medulloblastomas interact with the immune system in the brain. Furthermore, I highlight the translational value of preclinical models in the evaluation of new drug combinations ahead of clinical trial. Improving on the current tools available and accurately evaluating new therapies will ideally lead to improved clinical outcomes for patients with medulloblastoma.

Neuroendocrine tumors are a heterogeneous group of malignancies with an increasing prevalence. Since there is not much progress in therapy, model systems are urgently needed. We have a CEA424-SV40 TAg transgenic mouse model which develops spontaneous tumors in the antral region of the stomach. In addition, several cell lines derived from the tumor were established. Gene expression analysis of the tumor tissue as well as cell lines revealed neuroendocrine markers. Therefore we further characterized this model with special emphasis on the cells of origin and used it for testing new targeted treatment protocols. To analyze CEA424-SV40 TAg mouse model in more detail, tumor tissue as well as the cell lines derived from the primary tumor were investigated by immunohistochemistry, immunofluorescence, western blot, and ELISA. Antibodies used were directed at SV40 TAg, Ki-67, chromogranin A, chromogranin B, secretin, H+-K+-ATPase, glucagon, and transcription factors NeuroD1 and Nkx2.2. Plasma hormone levels of serotonin and secretin were measured by ELISA. Immunostainings of SV40 TAg and Ki-67 revealed highly proliferative tumors cells. The tumors stained intensively for the neuroendocrine markers chromogranin A, chromogranin B, secretin and glucagon. The tumor tissue as well as the cell lines expressed transcription factors NeuroD and Nkx2.2, which are involved in the differentiation of the neuroendocrine lineage. Hormone levels of serotonin and secretin in the plasma of the transgenic mice were dramatically elevated when compared with normal littermates, thus supporting the neuroendocrine phenotype. As the neuroendocrine phenotype of CEA424-SV40 TAg transgenic mouse was confirmed, molecularly targeted therapies were tested in this model system both in vitro and in vivo. Cell lines were tested for drug sensitivity with mTOR inhibitors (RAD001, NVP-BEZ235), paclitaxel, E2F inhibitor, HSP90 inhibitor, and p53 stabilizer Nutlin-3a. All the drugs tested in vitro could efficiently inhibit cell proliferation in a dose dependent manner. From these drugs the mTOR inhibitor RAD001 was chosen for the in vivo experiment. Daily feeding of 10 mg/kg RAD001 inhibited the tumor development and prolonged the survival time of the CEA424-SV40 TAg transgenic mice dramatically. The effects of the RAD001 treatment on tumor cells were achieved mainly through inactivating mTOR-p70S6K and mTOR-4EBP1 signaling as proven by western blot and immunohistochemistry. Still, some cells must develop escape mechanisms, since the tumor tend to grow. To gain a better understanding of the T antigen transforming mechanisms as well as the possible escape mechanisms, some efforts were made on the tumor originating cells in the CEA424-SV40 Tag transgenic mouse model. Possible candidates for these tumor originating cells in the stomach are the newly described epithelial as well as mesenchymal stem cells. In a first attempt, the expression feature of epithelial and mesenchymal stem cell markers were analyzed. Established cell lines as well as tumor tissue from the tumor bearing mice were investigated by reverse transcription PCR (RT-PCR), immunohistochemistry, immunofluorescence, western blot, and microarray analysis. From several markers analyzed, the tumor cell lines showed a high expression level of the potential epithelial stem cell marker Bmi1 in RT-PCR and cDNA expression array. This could be further substantiated by western-blotting and immunostaining. Consequently, Bmi1 message could also be found in the growing tumors both in mRNA and protein levels. Experiments using siRNA to knock down the SV40-TAg expression showed that the Bmi1 expression went down in the cell lines thus showing the interrelationship. On the other hand, the mesenchymal stem cell marker Etv1 was also found to be expressed in the tumor tissue and cell lines derived from the tumor. More interestingly, Etv1 expression level was up-regulated over the time course of the tumor development. From these, an Etv1 positive mesenchymal cell could be a possible candidate for transformation. Since the CEA-promoter used for the generation of the T-antigen transgenic animals contains Etv1 binding sites, it is tempting to speculate, that this may drive the transcription of the T antigen. In conclusion, our data provide convincing evidence that CEA424-SV40 TAg mice are a clinically relevant model for neuroendocrine tumor. Testing of molecularly targeted therapies both in vitro and in vivo offered promising candidates for further clinical evaluation. Thus, this new model system could be of great value not only for studies on the mechanisms of how SV40 TAg induces neuroendocrine tumors but also for exploring novel targeted therapy in a preclinical setting.<br>Neuroendokrine Tumore stellen heterogene Tumore mit ansteigender Prävalenz dar. Da es nur geringe Verbesserungen in der Therapie gibt, besteht ein Bedarf an neuen Modellsystemen. In unserer Arbeitsgruppe wurde das CEA424-SV40-T-Antigen transgene Mausmodell entwickelt, in welchem sich spontan Tumore im Bereich des Magenantrums bilden. Gleichzeitig wurden Zelllinien aus diesem Tumor etabliert. Die Analyse der Genexpression im Tumor und in den Zelllinien ergaben Hinweise auf einen neuroendokrinen Tumortyp. Deshalb wurde der Phänotyp des Tumors auch mit Blick auf den Ursprung der Tumore analysiert und für die Testung von Therapieoptionen eingesetzt. Die Analyse der CEA-424-SV40-TAg-Mäuse sowie der Tumorlinien wurden mittels Immunhistochemie, Immunfluoreszenz, Westernblot und ELISA untersucht. Dazu wurden Antikörper gegen das SV40-T-Antigen, den Proliferationsmarker Ki-67, Chromogranin A, Chromogranin B, Sekretin, die H-K-ATPase, Glukagon und die Transkriptionsfaktoren NeuroD und Nkx2.2 eingesetzt. Die Plasma Hormonkonzentrationen von Serotonin und Sekretin wurden mittels ELISA bestimmt. Die Färbung für das SV40-T-Antigen und Ki-67 zeigte einen hoch proliferierenden Tumor. Dieser war hoch positiv für die neuroendokrinen Marker Chromogranin A, Chromogranin B, Sekretin und Glukagon. Sowohl der Tumor als auch die Tumorzelllinien exprimierten die Transkriptionsfaktoren NeuroD und Nkx2.2, welche an der Differenzierung von neuroendokrinen Zellen beteiligt sind. Die Hormonkonzentrationen von Serotonin und Sekretin waren im Plasma der transgenen Mäuse deutlich erhöht. Dies ergab das Gesamtbild eines neuroendokrinen Karzinoms. In diesem Modell wurden nun verschiedene molekular begründete Therapien in vitro und in vivo getestet. So wurde an den Zelllinien die Empfindlichkeit von mTOR Inhibitoren (RAD001, NVP-BEZ235), Paclitaxel, E2F -Inhibitor, Hsp90-Inhibitor und dem p53 Inhibitor Nutlin3 getestet. Alle verwendeten Substanzen konnten die Proliferation der Tumorzellen dosisabhängig hemmen. Von diesen wurde dann der mTOR Inhibitor RAD001 für die in vivo Anwendung ausgewählt. RAD001 konnte dabei die Entwicklung der Tumore signifikant hemmen und verlängerte das Überleben der Tiere dramatisch. Der Effekt der mTOR Inhibition bestand dabei vor allem in der Hemmung des mTOR-p70S6K und mTOR-4EBP1 Pathways, was im Westernblot und der Immunhistochemie gezeigt werden konnte. Trotzdem muss festgehalten werden, dass einige Tumorzellen der Therapie entkommen konnten. Um nun Informationen über den Transformations- und den Escape-Mechanismus zu bekommen, wurde versucht die Tumorursprungszelle zu beschreiben. Mögliche Kandidaten dafür sind sowohl die jüngst beschriebenen intestinalen Epithelstammzellen als auch mesenchymale Stammzellen. Dazu wurden Marker-Gene an den etablierten Zelllinien und den Tumoren mit Hilfe von RT-PCR, Immunhistochemie, Immunfluoreszenz, Westernblot und Microarray untersucht. Dabei fand sich in den Tumorzellen eine hohe Expression des möglichen epithelialen Stammzellmarkers Bmi1. Dies konnte auch im Westernblot und in der Immunfärbung bestätigt werden. Folgerichtig fand sich dieser Marker auch in den wachsenden Tumoren. Experimente, in denen mit siRNA die Expression des SV40-T- Antigen blockiert wurde, ergaben eine Reduktion der Bmi1-Expression und weisen damit auf einen engen Zusammenhang hin. Gleichzeitig fand sich allerdings auch eine hohe Expression des mesenchymalen Stammzell-Markergenes Etv1 im Tumor und in den etablierten Zelllinien. Etv1 stieg dabei im Verlauf der Tumorentwicklung im Gewebe deutlich an. Deshalb könnte es sich bei der ursprünglich transformierten Zelle auch um eine mesenchymale Stammzelle handeln. Da der CEA Promotor, der die Expression des SV40-T-Antigens in den transgenen Mäusen regulieren soll, einige Bindungsstellen für den Transkriptionsfaktor Etv1 hat, liegt die Möglichkeit der Induktion des T-Antigens über Etv1 nahe. Dazu stehen aber noch weitere Experimente aus. Zusammenfassend zeigen die hier präsentierten Daten, dass es sich bei dem CEA424-SV40 Tag-Mausmodell um ein klinisch relevantes Modell für einen neuroendokrinen Tumor handelt. Zusammen mit den etablierten Tumorzelllinien können daran neue Therapieansätze getestet werden. Damit bietet es die Möglichkeit sowohl den Zusammenhang zwischen dem T-Antigen und der Entwicklung des neuroendokrinen Phänotyps als auch neue Therapieformen zu untersuchen.

Le cancer de l’ovaire constitue un véritable enjeu de santé public. Les nouveaux traitements se heurtent par ailleurs à des taux d’échec très élevés. Ceci s’explique notamment pour le manque de fiabilité des modèles précliniques classiques tels que la culture cellulaire en 2D. De nouveaux outils basés sur la culture cellulaire en 3D ont alors fait leur apparition tels que les sphéroïdes et les organoïdes. Or ces modèles ont leurs propres limites (coûts, difficultés d’application). La bio-impression 3D est une nouvelle approche permettant de créer des modèles tumoraux de manière contrôlée et reproductible. Néanmoins, elle a encore très peu été appliquée au cancer ovarien. En plus de la troisième dimension, il est important de prendre en compte les conditions dynamiques associées à l’environnement tumoral. Ceci est possible depuis quelques années grâce à la technologie des cancers sur puce basée sur la microfluidique. Cependant, cette technologie ne permet pas, à l’heure actuelle, de simuler le trajet vasculaire du médicament en amont de son interaction avec le tissu tumoral. Dans ce projet de thèse, nous avons souhaité créer un modèle tridimensionnel et dynamique du cancer ovarien en combinant les approches de bio-impression 3D et de microfluidique. Dans un premier temps, la bio-impression 3D a été utilisée pour créer la structure tumorale à proprement parlé. Pour y parvenir, nous avons formulé un hydrogel de gélatine et d’alginate de sodium dans lequel nous avons intégré des cellules cancéreuses ovariennes (SKOV-3) et des fibroblastes cancéreux (MeWo). Le tissu tumoral bio-imprimé a ensuite été caractérisé par différentes techniques pour démontrer sa viabilité et sa pertinence biologique. Sa réponse au cisplatine a également été évaluée. Dans un second temps, nous avons intégré le modèle tumoral bio-imprimé au sein d’un support microfluidique. Le rôle de ce support était de permettre la mise en culture du tissu bio-imprimé sous flux physiologique. Il devait également permettre de simuler le trajet vasculaire du médicament avant son interaction avec le tissu tumoral. Par la suite, nous avons fait appel à la simulation en mécanique des fluides pour concevoir une version améliorée du premier système. L’objectif étant de pouvoir tester, en même temps, plusieurs concentrations différentes de médicament sur un même dispositif microfluidique. Ce projet de thèse a démontré la capacité de la bio-impression 3D à créer des tissus tumoraux ovariens viables et fonctionnels. Il a par ailleurs ouvert des perspectives de recherche très intéressantes par rapport aux possibilités de combiner la bio-impression 3D et de la microfluidique en vue d’améliorer la modélisation préclinique des cancers ovariens<br>Ovarian cancer is a major public health issue. Moreover, new treatments still face very high failure rates. This is mainly due to the unreliability of conventional preclinical models such as 2D cell culture. Thus, new tools based on 3D cell culture have emerged such as spheroids and organoids. However, these models have their own limitations (cost, difficulty of application). 3D bioprinting is a new approach to create tunable and reproducible tumor models. However, very few bioprinted tumor models have been reported so far. Besides the “third dimension”, it is important to consider the dynamic conditions of the tumor environment. This has been possible for some years now thanks to microfluidics-based cancer-on-a-chip technology. However, this technology currently does not simulate the drug vascular transport before its interaction with the tumor cells. In this PhD project, we set out to create a dynamic, three-dimensional model of ovarian cancer by combining 3D bioprinting and microfluidics. First, 3D bioprinting was used to create the tumor structure itself. For that, we formulated a bio-ink comprising SKOV-3 ovarian cancer cells and MeWo cancer fibroblasts embedded in a gelatin – alginate hydrogel. The bioprinted tumor structures were then characterized by various techniques to demonstrate their viability and biological relevance. Their response to anticancer drug cisplatin was also assessed. In the second step, we integrated the bioprinted tumor model into a microfluidic support for culture under physiological flow. This support was also intended to simulate the drug's vascular transport prior to interaction with the tumor tissue. We then used computational fluid dynamics to design an improved version of the first system. The aim of this improved version was to simultaneously assess multiple drug concentrations. This PhD project demonstrated the ability of 3D bioprinting to create viable and functional ovarian tumor models. It has also brought interesting research prospects with regard to the possibilities of combining 3D bioprinting and microfluidics to improve preclinical modeling of ovarian tumors

Apoptosis is an important process in normal mammary gland physiology and evasion of apoptosis has also been identified as a hallmark of cancer. In breast cancer cells apoptotic resistance is an acquired feature that can promote tumour growth and progression. Induction of apoptosis by the extrinsic death ligand TRAIL has been shown to be a promising clinical therapy targeting a number of different cancer cells whilst sparing normal cells. Unfortunately most breast cancers are inherently resistant to TRAIL treatment. Herein it is shown that by reducing the expression of the downstream TRAIL inhibitor c-FLIP, a range of different breast cancer subtypes can be sensitised to TRAIL treatment resulting in significant cancer cell death. Significantly, suppression of c-FLIP in combination with TRAIL (FLIPi/TRAIL) ablated the tumour-initiating breast cancer stem cell (bCSC) subset, as defined by mammosphere formation assay, within cell lines. This selective killing of bCSCs translated to reduced tumour initiation and metastasis in animal transplant models. However, continued culture of FLIPi/TRAIL treated cell lines in adherent conditions resulted in bCSC re-acquisition suggesting a phenotypic plasticity of non-bCSC cells. Re-acquired bCSCs also demonstrated sensitivity to repeated FLIPi/TRAIL treatment and maintaining reduced c-FLIP expression prevented bCSC re-acquisition. These results substantiate the importance of resistance to apoptosis in tumour initiation and metastasis and identify the targeting of c-FLIP proteins as a promising anti-cancer therapeutic approach. Acquired resistance to existing mainstay therapies such as antiestrogens (AEs) (tamoxifen and Faslodex) and aromatase inhibitors (AIs) is an ongoing obstacle in treatment of a large number of breast cancer patients. AE-resistant models of breast cancer and a multiple endocrine-resistant patient sample demonstrated hypersensitivity to TRAIL. This sensitivity was observed in both in vitro and in vivo models of AE resistance and cell death was prevalent in both bulk tumour cells and bCSCs. Sensitisation was not attributed to combination AE/TRAIL treatment suggesting cellular changes during the acquisition of AE resistance are responsible for TRAIL sensitivity in these models. Further investigation suggested that the mechanism of AE-resistant cell sensitivity to TRAIL was not dependant on functional estrogen receptor signalling and is most likely dependant on the AE agent that the cancer cells have acquired resistance to. Interestingly tamoxifen-resistant MCF-7 cells were shown to have reduced c-FLIP protein expression compared to parental cells, further supporting c-FLIP’s potential in cancer therapy. Recent success in the use non-MHC-restricted γδ T cells as a targeted immunotherapy in clinical trials has identified this therapeutic methodology as desirable. Here it is shown that TRAIL is readily expressed by this subset of T cells that also demonstrate cytotoxicity to breast cancer cell lines. Neither the secretion of TRAIL or surface expression of TRAIL appeared to contribute significantly towards γδ T cell cytotoxicity and the majority of breast cancer cell death induced by γδ T cells would seem to be perforin-mediated. The suppression of c-FLIP in target cells increased γδ T cell cytotoxicity but again not via TRAIL. Preliminary results also indicated that the bCSCs of some cell lines were exquisitely sensitive to γδ T cell treatment. In summary these results indicate that targeting c-FLIP and TRAIL can be therapeutically beneficial in a range of different breast cancer subtypes by certain therapeutic strategies.

Oncolytic viruses promote anti-tumour immune response by direct tumour cell killing and activation of intratumoural immune system. The role of innate antiviral immune response to oncolytic viruses is still debated, as they counteract viral replication and trigger adaptive antitumor immunity. The DNA sensing-mediated cGAS/STING axis may act as a key balancer between lytic and immunotherapeutic activity of oncolytic viruses. Indeed, upon infection, viral DNA is sensed by cGAS/STING axis that, in turn, induces type-I interferon cascade counteracting viral replication and spread. For this reason, STING represents a hurdle for classical lytic-centric function of oncolytic viruses. On the other side, the immunological role of STING should also be considered, as it is emerging as a key bridge between innate and adaptive immunity. To evaluate the role of STING expression in tumour cells in response to onco-virotherapy, we generated murine STING KO tumour cell lines through CRISPR/Cas9 genome editing. Preclinical studies in syngeneic immunocompetent tumour-bearing mice showed that the inactivation of STING in tumour cells, while favouring oncolytic viral replication, impaired the immunotherapeutic effects of combination therapy based on herpetic oncolytic virus and PD1 blockade. Molecular characterization of tumours revealed that loss of STING prevents antitumour immune activation inducing a tolerogenic cell death and immunosuppressive tumour microenvironment. Accordingly, I propose that antiviral, tumourresident STING provides fundamental contributions to heat-up the TME eliciting immunotherapeutic efficacy of oncolytic viruses.

Tumour cell dissemination is a major issue with the treatment of cancer, thus new therapeutic strategies which can control this process are needed. Antagonism of integrins highly expressed in tumours is one potential strategy. The integrins are transmembrane glycoprotein adhesive receptors. Two of the integrins, αVβ3 and αIIbβ3, are highly expressed in a number of tumours and induce bi-directional signalling through their interaction with extracellular matrix proteins, and growth factor receptors. Through this signalling they play an important role in a number of cellular processes that are involved in tumour dissemination such as tumour growth, migration, invasion, metastasis and angiogenesis. Dual αIIbβ3 and αVβ3 integrin antagonism will have a direct effect on β3-expressing tumour cells that leads to the inhibition of cell migration and dissemination. Furthermore, through targeting tumour cell interaction with endothelial cells and platelets, this will also lead to inhibition of angiogenesis and metastasis. The aim of this project was to characterise the expression of αVβ3 and αIIbβ3 integrin in a panel of tumour cell lines and in human tumour xenograft samples, and to develop and utilise cell-based models to investigate potential novel β3 antagonists. The expression of αV and β3 subunits was detected in xenograft tissue using immunoblotting techniques. A panel of cell lines of different tumour types including melanoma, prostate, breast, colon and non small cell lung carcinoma was then characterised for αVβ3 and αIIbβ3 integrin expression using immunoblotting and immunocytochemistry. Melanoma cell lines demonstrated the strongest αVβ3 expression. No αIIbβ3 integrin expression was seen in any of the cell lines evaluated. A selection of cell lines with varying αVβ3 expression were then used to develop a functional test for cell migration, the scratch wound healing assay. Migration of tumour cells that expressed αVβ3 integrin was inhibited by the known β3 antagonists, cRGDfV peptide and LM609 antibody. A panel of 12 potential novel β3 integrin antagonists was screened for cytotoxicity and activity in the validated scratch assay. ICT9055 was the most effective antagonist in inhibition of M14 cell migration as determined by the scratch assay, with an IC₅₀ of < 0.1 μM. Therefore the work presented in this thesis has established models and tools for evaluating potential novel β3 integrin antagonists, and identified a promising molecule to progress for further preclinical evaluation.

Despite significant advances in cancer treatment, clinical response remains suboptimal and there is a continued requirement for improved chemotherapeutics. The attrition rate for new therapies is high, due principally to lack of in vivo efficacy and poor pharmacodynamics. Consequently better systems are required to determine in vivo preclinical efficiency and drug-target interactions. Engineering of cancer cells to express fluorescent or bioluminescent proteins, either endogenously or under the control of specific gene promoters, and their detection by noninvasive optical imaging has the potential to improve preclinical drug development. In this study, a panel of colorectal cancer cell lines were engineered to express fluorescent and luminescent proteins either constitutively or under control of gene-promoters for the DNA damage response gene p53 or the cell cycle regulator p21, both important pharmacodynamic sensors. These cell lines were characterised for their potential as in vivo models of primary and metastatic tumour therapy response, several showing significant potential. In addition to the development of these models, this study also addressed the pharmacokinetics of different luciferase substrates and identified optimal temporal and dose characteristics for each. Furthermore, a new application for bioluminescent imaging was developed and validated for use in preclinical evaluation of vascular disrupting agents, a new generation of cancer therapeutic. This study demonstrates that despite the dynamic and variable nature of fluorescent and bioluminescent imaging, reproducible results can be obtained if appropriate precautions are taken. The models developed herein will expedite cancer drug development whilst reducing and refining the use of animals in research.

Bien que la radiothérapie, traitement incontournable en neuro-oncologie, améliore la survie des patients, elle affecte de manière considérable le tissu cérébral sain avoisinant la tumeur conduisant à des déficits cognitifs qui sont retrouvés chez 50 à 90 % des patients. Les avancées technologiques réalisées lors de la dernière décennie ont permis de concevoir de nouvelles techniques d’irradiation avec des propriétés balistiques prometteuses. Cependant, leurs intérêts pour prévenir la radiotoxicité cérébrale reste à démontrer, en s’appuyant notamment sur des recherches précliniques pour lesquelles l’utilisation de ces techniques de radiothérapie est fragmentaire à ce jour. L’objectif de ces travaux de thèse a été de caractériser, chez le rat adulte sain ou porteur d’une tumeur cérébrale, les effets de l’irradiation cérébrale ciblée sur l’intégrité tissulaire et les déficits cognitifs, d’une part de manière multiparamétrique via l’utilisation de l’imagerie IRM, de différents tests comportementaux ainsi que des analyses immunohistologiques, et d’autre part de manière longitudinale avec un suivi des animaux jusqu’à 6 mois après irradiation. Collectivement, nos données montrent, comme attendu et en accord avec la littérature, que l’irradiation du cerveau sain entier engendre des déficits dans les processus d’apprentissage, de mémorisation et d’émotion, et ceci pendant les phases aiguës et chroniques. De même, ce paradigme d’irradiation est associé à des altérations du tissu cérébral. Cependant, et d’une manière un peu surprenante par rapport à notre hypothèse de départ, l’irradiation d’un seul hémisphère n’a pas modifié de façon significative les performances cognitives évaluées et n’a pas altéré l’intégrité du tissu cérébral. Les résultats obtenus dans le modèle de tumeur cérébrale montrent des déficits cognitifs suite à une irradiation cerveau entier, lesquels sont aussi observés avec l’irradiation hémisphérique mais avec des effets moindres. Malheureusement, du fait des effectifs faibles au sein des groupes expérimentaux, il est difficile de conclure sur le fait que les déficits cognitifs radio-induits observés soient exacerbés en présence de tumeur<br>Although radiotherapy, an essential treatment in neuro-oncology, improves the survival of patients, it significantly affects the surrounding healthy brain tissue, leading to cognitive deficits found in 50 to 90% of patients. Technological advancements made in the last decade have allowed the development of new irradiation techniques with promising ballistic properties. However, their potential for preventing cerebral radiotoxicity remains to be demonstrated, relying mainly on preclinical research, for which the use of these radiotherapy techniques is currently fragmented. The objective of this thesis work was to characterize the effects of targeted brain irradiation on tissue integrity and cognitive deficits in healthy adult rats and rats bearing brain tumor. This characterization was done through multiparametric imaging using MRI, various behavioral tests, as well as immunohistological analyses. Furthermore, a longitudinal approach was employed, with the animals being monitored up to 6 months after irradiation. Collectively, our data demonstrate, as expected and in accordance with the literature, that whole-brain irradiation leads to deficits in learning, memory, and emotion processes, both during acute and chronic phases. Similarly, this irradiation paradigm is associated with alterations in brain tissue. However, somewhat surprisingly compared to our initial hypothesis, irradiation of a single hemisphere did not significantly modify the evaluated cognitive performances or compromise tissue integrity. In the brain tumor model, cognitive deficits were observed following whole-brain irradiation, which were also present with hemispheric irradiation but with lesser effects. Unfortunately, due to low sample sizes within the experimental groups, it is difficult to conclude whether the observed radio-induced cognitive deficits are exacerbated in the presence of a tumor

Despite significant advances in cancer treatment, clinical response remains suboptimal and there is a continued requirement for improved chemotherapeutics. The attrition rate for new therapies is high, due principally to lack of in vivo efficacy and poor pharmacodynamics. Consequently better systems are required to determine in vivo preclinical efficiency and drug-target interactions. Engineering of cancer cells to express fluorescent or bioluminescent proteins, either endogenously or under the control of specific gene promoters, and their detection by noninvasive optical imaging has the potential to improve preclinical drug development. In this study, a panel of colorectal cancer cell lines were engineered to express fluorescent and luminescent proteins either constitutively or under control of gene-promoters for the DNA damage response gene p53 or the cell cycle regulator p21, both important pharmacodynamic sensors. These cell lines were characterised for their potential as in vivo models of primary and metastatic tumour therapy response, several showing significant potential. In addition to the development of these models, this study also addressed the pharmacokinetics of different luciferase substrates and identified optimal temporal and dose characteristics for each. Furthermore, a new application for bioluminescent imaging was developed and validated for use in preclinical evaluation of vascular disrupting agents, a new generation of cancer therapeutic. This study demonstrates that despite the dynamic and variable nature of fluorescent and bioluminescent imaging, reproducible results can be obtained if appropriate precautions are taken. The models developed herein will expedite cancer drug development whilst reducing and refining the use of animals in research.

Les progrès de la recherche sur le cancer du sein dépendent de la disponibilité d’outils appropriés, comme les lignées cellulaires qui peuvent être implantées chez des souris immunocompétentes. La souche de souris C57Bl/6 est la plus étudiée et c’est la seule pour laquelle certaines variantes génétiques sont disponibles. Étant donné qu'aucune lignée cellulaire de carcinome mammaire à récepteurs hormonaux positifs de souche C57Bl/6 n'est disponible, nous avons décidé d'établir des lignées cellulaires de ce type. Nous avons induit des cancers du sein chez des souris C57BL/6 femelles en utilisant un analogue synthétique de la progestérone combiné à un agent endommageant l'ADN. Des lignées cellulaires ont été établies à partir de ces tumeurs et sélectionnées pour leur positivité au niveau du double récepteur (estrogène + progestérone), ainsi que pour leur transplantabilité chez les femelles C57BL/6. Parmi plusieurs lignées, une lignée cellulaire, que nous avons appelée MD5, remplissait ces critères et a permis l'établissement de tumeurs mal différenciées et très prolifératives. Ces tumeurs ont réduit leur croissance (sans toutefois régresser) lors du traitement par des antagonistes des récepteurs d’œstrogènes, ainsi que par une chimiothérapie à base d'anthracylines. Cependant, ce dernier effet n'a pas été influencé par la déplétion des lymphocytes T et, en outre, ces tumeurs n'ont pas répondu au blocage de PD-1, ce qui suggère que les tumeurs MD5 sont immunologiquement froides. En conclusion, les cellules MD5, dérivées des animaux C57BL/6, constituent un modèle de cancer du sein à récepteurs hormonaux positifs de mauvais pronostic<br>Progress in breast cancer research relies on the availability of suitable cell lines that can be implanted in immunocompetent laboratory mice. The best explored mouse strain, C57Bl/6, is also the only one for which multiple genetic variants are available. Driven by the fact that no hormone receptor-positive C57Bl/6-derived mammary carcinoma cell lines are available, we decided to establish such cell lines. Breast cancers were induced in female C57BL/6 mice using a synthetic progesterone analogue combined with a DNA damaging agent. Cell lines were established from these tumors and selected for dual (estrogen + progesterone) receptor positivity, as well as transplantability into C57BL/6 females. One cell line, which we called MD5,fulfilled these criteria and allowed for the establishment of poorly differentiated, highly proliferative, immune cold tumors. Such tumors reduced their growth (though did not regress) upon treatment with estrogen receptor antagonists, as well as with anthracyline-based chemotherapy. However, the latter effect was not influenced by T cell depletion and MD tumors failed to respond to PD-1 blockade, suggesting that they are immunologically cold. In conclusion, C57BL/6-derived MD5 cells constitute a model of poor prognosis hormone receptor-positive breast cancer

Pancreatic ductal adenocarcinoma (PDAC) is a disease with one of the highest mortality rates and with an increasing incidence worldwide. Currently, clinically administered therapies for the PDAC treatment are still extremely ineffective and of limited access. Given this scenario, it is urgent to investigate and validate new therapies for the treatment of this neoplasia. PDAC is a cancer with a unique stratified bio-architecture and characterized by an exacerbated desmoplastic reaction involving cancer-associated fibroblasts, immune cells and extracellular matrix proteins (ECM), which play a significant role in tumor progression and resistance to the currently used therapies in a clinical setting. The absence of cell-based in vitro models capable of reproducing the PDAC desmoplastic microenvironment and the histo-morphology results in a low correlation between the performance of new therapies in preclinical trials and that observed in controlled clinical trials. In this sense, three-dimensional (3D) in vitro tumor models emerge as a more suitable solution for preclinical evaluation when compared with the most frequently used two-dimensional (2D) cell cultures. 3D models represent more biomimetic models as they allow a more robust and realistic recapitulation of the tumor microenvironment, contributing to the discovery of new biomarkers and to the pre-clinical evaluation of new drugs in a more accurate way. Amongst currently developed 3D in vitro PDAC platforms, few are those that accurately recapitulate cell heterogeneity, tumor architecture and fibrotic stroma. To overcome these limitations, this dissertation focuses on bioengineering and characterization of a new 3D PDAC model, consisting of a biomimetic co-culture of pancreatic cancer cells (PANC-1) and cancer-associated fibroblasts (CAFs). This 3D model demonstrated to recapitulate the cellular components, their spatial distribution and resistance to pharmacological therapies in a similar way to that found in human tumors.<br>O adenocarcinoma ductal pancreático (ADP) é uma doença com uma das maiores taxas de mortalidade e com uma incidência crescente a nível mundial. Atualmente, as terapias administradas na clínica para o tratamento do ADP são ainda extremamente ineficazes e de acesso limitado. Perante este cenário torna-se urgente a investigação e validação de novas terapias para o tratamento desta neoplasia. O ADP é um cancro com uma bioarquitetura estratificada única e caracterizado por uma exacerbada reação desmoplásica envolvendo fibroblastos associados ao cancro, células imunes e proteínas da matriz extracelular (MEC), que desempenham um papel significativo na progressão tumoral e na resistência às terapias utilizadas atualmente em contexto clínico. A ausência de modelos de celulares capazes de reproduzir in vitro o microambiente desmoplásico e a histo-morfologia do ADP origina uma baixa correlação entre a performance de novas terapias obtida em ensaios pré-clínicos e aquela observada em ensaios clínicos controlados. Neste sentido, os modelos de tumores tridimensionais (3D) in vitro surgem como uma solução mais adequada para a avaliação pré-clínica quando comparados com as recomendadas culturas celulares bidimensionais 2D. Os modelos 3D representam modelos mais biomiméticos pois permitem recapitular de uma forma mais robusta e realista o microambiente tumoral, contribuindo para a descoberta de novos biomarcadores e para a avaliação pré-clínica de novos fármacos de uma forma mais precisa. Das plataformas 3D in vitro de ADP desenvolvidas atualmente, poucas são as que recapitulam de forma precisa a heterogeneidade celular, a arquitetura tumoral e o estroma fibrótico. Com o objetivo de colmatar estas limitações, a presente dissertação foca na bioengenharia e caracterização de um novo modelo 3D de ADP, consistindo numa co-cultura biomimética de células cancerígenas pancreáticas (PANC-1) e fibroblastos associados ao cancro (FACs). Este modelo 3D demonstrou recapitular os componentes celulares, a sua distribuição espacial e a resistência a terapias farmacológicas de uma forma semelhante à encontrada nos tumores humanos.<br>Mestrado em Biotecnologia

Introduction: One of the main obstacle to the successful development of therapeutic strategies remains the identification of biomarker underlying drug resistance. Recently, investigators have become more aware the role of the tumor microenvironment (TME) in cancer and the potential therapeutic opportunities that derive from suppressing potential resistance mechanisms arising microenvironmental interactions. The aim of this study was to set-up multicellular culture models to uncover the molecular mechanisms by which stromal/endothelial cells modulate response to signaling inhibitors and to identify potential therapeutic targets in PTEN-loss contexts. Methods and Materials: Isogenic CRC cell lines (X-MAN™ HCT116 and HCT116 PTEN-/-) were treated with MAPKi and PI3K/mTORi alone or in combination, in the presence or absence of stromal fibroblasts or fibroblast/endothelial cell conditioned medium (CM). Cytofluorimetric analysis and Crystal Violet assay were used to analyse functional response to targeted agents; pathways activation and cytokine/chemokine profile were analysed using Western blot and ELISA assay respectively. Results and Discussion: In co-culture CRC models, the response to MAPK and PI3K inhibitors is the result of interaction between tumor cells and their surrounding stroma. The response to PI3K/mTORi is mainly influenced by microenvironmental interactions: direct cell-to-cell tumor/stroma contact renders stromal cells resistant to PI3K/mTORi, while the presence of stromal cell-derived soluble factors sensitizes PTEN-competent CRC cells to PI3K/mTORi-mediated growth inhibition. This effect was confirmed using CM from different types of stromal cells (fibroblast/endothelial) that similarly affected the response of CRC cell lines to signalling inhibitors; this is probably due to similar profile of cytokine/chemokine production in stromal cell and is subjected to a “saturation” effect. The presence of stromal CM upregulates MAPK activation regardless of PTEN status, whereas mTOR pathway upregulation is observed mainly in PTEN-competent CRC cellsin PTEN-competent cells soluble factors released by stromal elements paradoxically impair PTEN function, leading to downstream mTORC1 complex formation and pathway activation. This paradoxical mTORC1 activation upon exposure to stroma-derived soluble factors results in functional hypersensitivity of PTEN-competent CRC cells to the growth inhibitory effects of double PI3K/mTOR inhibitors. . Conclusions: The presence of stromal cells (fibroblasts/endothelium) profoundly influences CRC response to PI3K/mTOR-targeting agents. Understanding the mechanisms underlying microenvironmental interactions (tumor, stroma, soluble factors) may be of fundamental importance to overcome therapeutic resistance and develop more effective therapies for patients affected by cancer.

Yes<br>Non-invasive methods to monitor tumour growth are an important goal in cancer drug development. Thermographic imaging systems offer potential in this area, since a change in temperature is known to be induced due to changes within the tumour microenvironment. This study demonstrates that this imaging modality can be applied to a broad range of tumour xenografts and also, for the first time, the methodology’s suitability to assess anti-cancer agent efficacy. Mice bearing subcutaneously implanted H460 lung cancer xenografts were treated with a novel vascular disrupting agent, ICT-2552, and the cytotoxin doxorubicin. The effects on tumour temperature were assessed using thermographic imaging over the first 6 hours post-administration and subsequently a further 7 days. For ICT-2552 a significant initial temperature drop was observed, whilst for both agents a significant temperature drop was seen compared to controls over the longer time period. Thus thermographic imaging can detect functional differences (manifesting as temperature reductions) in the tumour response to these anti-cancer agents compared to controls. Importantly, these effects can be detected in the first few hours following treatment and therefore the tumour is observable non-invasively. As discussed, this technique will have considerable 3Rs benefits in terms of reduction and refinement of animal use.<br>University of Bradford

Les tumeurs des cellules de la granulosa (GCTs) sont des tumeurs avec un potentiel malin ayant tendance à récidiver, provoquant ainsi la mort dans 80% des cas de stade avancé consécutif à une rechute. Bien que les GCTs représentent 5% des tumeurs ovariennes, peu d’études ont évalué les protocoles de traitement adjuvant pour la maladie avancée ou récurrente. Notre but était d’évaluer l’efficacité de la voie de signalisation du facteur de croissance de l’endothélium vasculaire A (VEGFA) comme cible pour le traitement de la GCT utilisant le modèle murin transgénique Ptentm1Hwu/tm1Hwu; Ctnnb1tm1Mmt/+; Amhr2tm3(cre)Bhr/+ (PCA) qui reproduit le stade avancé de la maladie humaine. Un anticorps anti-VEGFA a été administré une fois par semaine par voie intrapéritonéale (IP) à partir de 3 semaines d’âge. La thérapie anti-VEGFA a permis une réduction de la taille des tumeurs à 6 semaines d’âge (p<0.05) et une prolongation de la survie des animaux traités, lorsque comparé aux animaux contrôles. L’analyse des GCTs a montré une réduction significative de la prolifération cellulaire (p<0.05) et de la densité microvasculaire (p<0.01) mais aucune différence significative n’a été détectée dans l’apoptose cellulaire. p44/p42 MAPK, un effecteur de la signalisation pour le récepteur 2 de VEGFA (VEGFR2) associé à la prolifération cellulaire, était moins activé dans les tumeurs traitées (p<0.05). Par contre, l’activation d’AKT, un effecteur impliqué dans la survie cellulaire, était similaire d’un groupe à l’autre. Ces résultats suggèrent que l’anticorps anti-VEGFA réduit la prolifération cellulaire et la densité microvasculaire chez les souris PCA par inhibition de la voie de signalisation VEGFR2-MAPK, inhibant ainsi la croissance tumorale. En conclusion, l’efficacité de la thérapie anti- VEGFA mérite d’être évaluée en essais contrôlés randomisés pour le traitement des GCTs chez l’homme.<br>Ovarian granulosa cell tumors (GCTs) are potentially malignant tumors that have a tendency for late recurrence and cause death in 80% of women with advanced GCT due to recurrent disease. Although GCTs represent 5% of ovarian tumors in women, few studies have evaluated adjuvant treatment protocols for advanced or recurrent disease. Our goal was to determine the potential of targeting the vascular endothelial growth factor A (VEGFA) signaling pathway for the treatment of GCT. We used a genetically engineered mouse model, Ptentm1Hwu/tm1Hwu; Ctnnb1tm1Mmt/+; Amhr2tm3(cre)Bhr/+ (PCA), which imitates the advanced human disease. A monoclonal anti-VEGFA antibody was administered by intra-peritoneal injection once a week beginning at 3 weeks of age. Anti-VEGFA therapy significantly decreased tumor weights by 6 weeks of age (p<0.05) and increased survival in treated animals in comparison to controls. Significant decreases in tumor cell proliferation (p<0.05) and microvessel density (p<0.01), but no significant difference in apoptosis was found in PCA tumors. p44/p42 MAPK, a VEGFA receptor 2 (VEGFR2) signaling effector associated with cell proliferation, was significantly less activated in anti-VEGFA-treated tumors (p<0.05). In contrast, AKT activation, a VEGFR2 signaling effector associated with cell survival was similar among all groups. These results suggest that anti-VEGFA therapy effectively reduces cell proliferation and microvessel density in PCA mice by inhibition of the VEGFR2-MAPK pathway, resulting in inhibition of GCT growth. We conclude that anti-VEGFA therapy merits further investigation in the form of controlled randomized trials for the treatment of human GCT.

Gliomas, encompassing Low Grade (LGGs) and High Grade (HGGs) diseases are, overall, the most common solid tumors of the pediatric population. Children affected by LGGs that cannot be cured often struggle to obtain a complete remission of the disease, suffering from significant long-term sequelae. HGGs are instead associated with very grim outcomes and no innovative treatment so far has been able to change this poor prognosis. New treatment approaches, such as immunotherapy, are therefore needed for these patients. Gliomas, as all solid tumors, are characterized by a high structural complexity and a strongly immunosuppressive microenvironment, both responsible for the frequent failure of immunotherapeutic approaches. Such complexity is difficult to reproduce with common bidimensional (2D) systems, making the conventional tumor models unable to predict the real efficacy of these treatments. In order to overcome such hurdles and obtain more reliable tumor models, we developed an innovative fibrin-based hydrogel 3D model and tested it as platform to establish primary cultures of pediatric LGGs and HGGs. In particular, LGGs are otherwise extremely difficult to maintain in culture owing to the well-known activation of senescence pathways. To date, 37 samples of LGG were cultured in both 2D and 3D platforms, showing that the 3D-culture enables the stabilization of LGG. Cell lines identity was confirmed by short tandem repeats (STRs) and the immunohistochemical characterization (structure, H&amp;E, Ki67, tumor and differentiation markers) in 3D cultures revealed phenotype and cellular organization closer to those observed in the onset sample, as compared to 2D. The 3D-cultured tumor cells showed a significantly (p =0,008) lower senescence rate than 2D. Furthermore, also 6 samples of HGG have been cultured and established as well. We therefore evaluated the antitumor activity of an innovative immunotherapeutic approach based on the use of γδ-T cells. These cells display properties of both the innate and the adaptive compartment, have a powerful cytotoxic activity, are able to recognize antigens independently from the HLA and display a negligible alloreactivity, resulting extremely attractive for clinical translation in particular for the establishment of an allogeneic “third-party T-cell bank” for ready for use off-the-shelf adoptive T-cell product. The clinical exploitation is limited, however, by the low percentage of circulating γδ-T cells and the lack, to date, of effective culture systems to expand them. We have developed a protocol to expand a clinically relevant number of polyclonal memory γδ-T cells. Artificial antigen presenting cells (aAPC) expressing CD86/4-1BBL/CD40L and cytomegalovirus-pp65 antigen were used to induce γδ-T cell expansion and activation. For increase the safety of our approach aAPCs has been further stable gene-modified with a suicide gene, inducible caspase 9, and then single cell cloned. The γδ-T cells expanded and expressed markers of activation and memory, maintaining a polyclonal phenotype (with a predominantly Vδ1 well known memory infiltrating population), negligible alloreactivity and potent antitumor capacity versus a broad range of malignancies. Lastly, we characterized the antitumor activity of expanded γδ-T cells, as well as of more conventional approaches (such as radiotherapy and chemotherapy), in 2D and in 3D-cultured tumors, showing that the first are significantly more sensitive to treatment. Expanded γδ-T cells showed a relevant antitumor activity in both 2D and 3D, the latter requiring longer cultures to obtain an antitumor control. These data suggest a more realistic estimation of the efficacy of such treatment by the 3D-platform, confirming it as a better model for biological and therapeutic studies.