Η διατριβή αφορά την ανάπτυξη και υλοποίηση ενός τετραδιάστατου, διακριτού μοντέλου προσομοίωσης της συμπεριφοράς καρκινικών κυτταρικών συστημάτων σε ελεύθερη ανάπτυξη και της απόκρισής τους σε χημειοθεραπευτική ή και ακτινοθεραπευτική αγωγή. Υλοποιήθηκαν δύο εκδοχές του μοντέλου: η χωρική και η μη χωρική προσέγγιση. Η χωρική προσέγγιση αναφέρεται στην τετραδιάστατη προσομοίωση συμπαγών όγκων. Η μη χωρική προσέγγιση βρίσκει εφαρμογή στην περίπτωση μη συμπαγών όγκων, καθώς και συμπαγών όγκων, όταν δεν δίνεται έμφαση στη χωρική εξέλιξή τους. Η ερευνητική εργασία έχει επικεντρωθεί σε τρεις τύπους καρκινικών όγκων: καρκίνος του μαστού, καρκίνος του πνεύμονα και πολύμορφο γλοιοβλάστωμα και σε θεραπευτικά σχήματα χορήγησης των σκευασμάτων: επιρουβικίνη (epirubicin), τεμοζολομίδη (temozolomide), σισπλατίνη (cisplatin), γεμσιταμπίνη (gemcitabine), βινορελμπίνη (vinorelbine) και δοσεταξέλη (docetaxel). Σκοπός της εργασίας είναι η ανάπτυξη ενός εργαλείου για την αξιόπιστη υποστήριξη ιατρών στη λήψη αποφάσεων σχετικά με την επιλογή θεραπευτικών σχημάτων και την εξατομικευμένη βελτιστοποίηση της θεραπευτικής αγωγής. Η αφετηρία είναι η μοντελοποίηση του κυτταρικού κύκλου και των πιθανών μεταβάσεων μεταξύ των καταστάσεων που μπορεί να βρεθεί ένα κύτταρο. Το μοντέλο βασίζεται στην υπόθεση ότι ο καρκινικός όγκος διατηρείται από μια συγκεκριμένη κατηγορία κυττάρων, τα καρκινικά βλαστικά κύτταρα (cancer stem cells), και έχει επεκταθεί ώστε να περιλαμβάνει σε μεγαλύτερη λεπτομέρεια διάφορους βιολογικούς μηχανισμούς σε μοριακό (πχ. εκφράσεις γονιδίων) και κυτταρικό επίπεδο. Ο μηχανισμός δράσης, η φαρμακοκινητική και η φαρμακοδυναμική των θεωρούμενων σκευασμάτων έχουν μελετηθεί βιβλιογραφικά και έχουν ενσωματωθεί στο μοντέλο. Επίσης, το μοντέλο έχει αναπτυχθεί ώστε να λαμβάνει υπόψη του την κλινική εικόνα του ασθενούς με χρήση εξατομικευμένων κλινικών δεδομένων, όπως απεικονιστικά δεδομένα (π.χ. CT, MRI, PET), ιστοπαθολογικά δεδομένα (π.χ. τύπος όγκου, βαθμός διαφοροποίησης) και μοριακά δεδομένα (π.χ. έκφραση γονιδίων). Στα πλαίσια της διατριβής πραγματοποιούνται έλεγχοι αξιοπιστίας και εκτενείς παραμετρικές μελέτες για την αποσαφήνιση της ευαισθησίας του μοντέλου στη διακύμανση των παραμέτρων του τόσο κατά την προσομοίωση της ελεύθερης ανάπτυξης όσο και κατά την εφαρμογή της χημειοθεραπευτικής αγωγής. Η ποσοτική αξιολόγηση, προσαρμογή και βελτιστοποίηση του μοντέλου πραγματοποιείται στα πλαίσια των ευρωπαϊκών ερευνητικών προγραμμάτων ACGT (Advancing Clinicogenomic Trials on Cancer, FP6-2005-IST-026996), ContraCancrum (Clinically Oriented Cancer Multilevel Modelling, FP7-ICT-2007-2-223979) και P-medicine (From data sharing and integration via VPH models to Personalized medicine, FP7-ICT-2009-6-270089) μέσω της αξιοποίησης πραγματικών κλινικών δεδομένων. Στην παρούσα διατριβή παρουσιάζονται τα αποτελέσματα της προσαρμογής του μοντέλου σε κλινικά δεδομένα του καρκίνου του μαστού, του καρκίνου του πνεύμονα και του πολύμορφου γλοιοβλαστώματος. Επιπλέον, διάφορες εκδόσεις του μοντέλου έχουν αξιοποιηθεί για ‘την επάνδρωση’ μιας ευρωπαϊκής βάσης μοντέλων για τον καρκίνο, που υλοποιείται στα πλαίσια του ευρωπαϊκού ερευνητικού προγράμματος TUMOR (Transatlantic Tumour Model Repositories, FP7-ICT-2009-5-247754). Το μοντέλο υλοποιείται σε γλώσσα προγραμματισμού C++. / In the present thesis, a clinically oriented, multiscale, discrete simulation model of cancer free growth and response to chemotherapy and/or radiotherapy is presented and investigated. Two versions of the model have been implemented: the spatial and the non spatial approach. The spatial model concerns the spatiotemporal evolution of solid tumours, whereas the non spatial model can be applied in the case of non solid cancers, as well as solid tumours, when no emphasis is put on the spatial features of a tumour evolution. The research work has been focused on the paradigms of early breast cancer treated with the single agent epirubicin, primary lung cancer treated with various combinations of cisplatin, gemcitabine, vinorelbin and docetaxel and glioblastoma multiforme treated with combined modality treatment using radiation and chemotherapy with temozolomide. The goal is to end up with a reliable simulation system able to assist clinicians in selecting the most appropriate therapeutic pattern, extracted from several candidate therapeutic schemes in the context of patient individualized treatment optimization. The model incorporates the biological mechanisms of cell cycling, quiescence, recruitment (reentry into the cell cycle), differentiation and death. It is based on the well documented assumption that tumour sustenance is due to the existence of cancer stem cells, i.e. cells which have the ability to preserve their own population, as well as give birth to cells that follow the path towards terminal differentiation. Furthermore, the mechanism of action, pharmacokinetics and pharmacodynamics of all considered agents have been bibliographically studied and incorporated into the model. Finally, the model has been developed to support and incorporate individualized clinical data such as imaging data (e.g. CT, MRI, PET slices, possibly fused), including the definition of the tumour contour and internal tumour regions (proliferating, necrotic), histopathologic (e.g., type of tumour) and genetic data (e.g., gene expression). An exhaustive and in-depth examination of the model behaviour with respect to the variation of its input parameters has been performed, in order to determine the impact of its parameters, guarantee a biologically relevant virtual tumour behaviour and enlighten aspects of the interplay and possible interdependencies of the biological mechanisms modeled. Finally, the model has been quantitativily validated and adaptated in the framework of the ACGT (Advancing Clinicogenomic Trials on Cancer, FP6-2005-IST-026996), ContraCancrum (Clinically Oriented Cancer Multilevel Modelling, FP7-ICT-2007-2-223979) and P-medicine (From data sharing and integration via VPH models to Personalized medicine, FP7-ICT-2009-6-270089) European Commission-funded projects by exploiting real clinical data. In the present thesis, the clinical adaptation of the model focuses on breast cancer, lung cancer and glioblastoma multiforme clinical cases. Moreover, various versions of the model have been uploaded to the EU cancer model repository developed by the TUMOR (Transatlantic Tumour Model Repositories, FP7-ICT-2009-5-247754) European Commission-funded project. The model has been developed in the C++ programming language.

Liu, Pi-chu. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2011. / Includes bibliographical references (leaves 155-173). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.

Chan Yiu Leung. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2006. / Includes bibliographical references (leaves 123-149). / Abstracts in English and Chinese. / Abstracts --- p.I / Acknowledgements --- p.VIII / List of Figures --- p.IX / Lists of Abbreviations --- p.X / Contents --- p.XII / Chapter Chapter One: --- Introduction and Literature Review --- p.1 / Chapter 1.1 --- Cancer in General --- p.1 / Chapter 1.2 --- Glioma --- p.3 / Chapter 1.2.1 --- Types of Glioma --- p.6 / Chapter 1.2.1.1 --- Astrocytomas --- p.6 / Chapter 1.2.1.2 --- Oligodendrogliomas --- p.8 / Chapter 1.2.1.3 --- Ependymomas --- p.9 / Chapter 1.2.2 --- Glioblastoma Multiforme (GBM) --- p.10 / Chapter 1.2.3 --- Molecular Biology of GBM --- p.11 / Chapter 1.2.4 --- Current Treatment for GBM --- p.15 / Chapter 1.3 --- HMG-Co A reductase inhibitors --- p.17 / Chapter 1.3.1 --- Pharmacology of HMG-Co A reductase inhibitor --- p.18 / Chapter 1.3.2 --- Epidemiological link between HMG-Co A Reductase Inhibitors and Cancer --- p.20 / Chapter 1.3.3 --- Current HMG-Co A reductase inhibitors research in cancer --- p.21 / Chapter 1.3.3.1 --- Inhibition of tumor cell growth --- p.21 / Chapter 1.3.3.2 --- Inhibition of Angiogenesis --- p.22 / Chapter 1.3.3.3 --- Anti-invasive effects of HMG-Co A reductase inhibitors.… --- p.23 / Chapter 1.3.3.4 --- Apoptosis induction by HMG-Co A reductase inhibitors --- p.24 / Chapter 1.3.4 --- In vivo efficacy and synergistic effects --- p.25 / Chapter 1.4 --- Tumor Necrosis Factor (TNF) related apoptosis-inducing Ligand (TRAIL) --- p.28 / Chapter 1.4.1 --- Molecular mechanisms of TRAIL-induced apoptosis --- p.29 / Chapter 1.4.2 --- Role for TRAIL in cancer therapy --- p.30 / Chapter 1.5 --- Objectives --- p.34 / Chapter Chapter 2 --- Methods and Materials --- p.35 / Chapter 2.1 --- Cell culture --- p.35 / Chapter 2.2 --- Cell proliferation detection (MTT) methods --- p.36 / Chapter 2.3 --- "Caspase 3,9 activities induced by lovastatin" --- p.37 / Chapter 2.4 --- Detection of apoptosis by Annexin V and PI staining --- p.39 / Chapter 2.5 --- Cell cycle analysis protocols --- p.41 / Chapter 2.6 --- DNA fragmentation ELISA detection kit protocols --- p.42 / Chapter 2.7 --- Reverse Transcription (RT) Polymerase Chain Reaction (PCR) --- p.44 / Chapter 2.8 --- Polymerase Chain Reaction (PCR) --- p.46 / Chapter 2.9 --- Bio-molecules extraction/purification protocols --- p.48 / Chapter 2.10 --- "Microarray analysis on lovastatin treated glioblastoma cells A172, M059J and M059K" --- p.51 / Chapter 2.10.1 --- Cells treatment and RNA extraction --- p.51 / Chapter 2.10.2 --- Synthesis of first strand cDNA --- p.53 / Chapter 2.10.3 --- Synthesis of second strand cDNA --- p.54 / Chapter 2.10.4 --- Purification of double stranded cDNA --- p.54 / Chapter 2.10.5 --- Synthesis of cRNA by in vitro transcription (IVT) --- p.55 / Chapter 2.10.6 --- Recovery of biotin-labelled cDNA --- p.56 / Chapter 2.10.7 --- Fragmentation of cRNA --- p.56 / Chapter 2.10.8 --- Preparation of hybridization reaction mixtures --- p.57 / Chapter 2.10.9 --- Loading of reaction mixtures into bioarray chambers --- p.58 / Chapter 2.10.10 --- Hybridization --- p.58 / Chapter 2.10.11 --- Post-hybridization wash --- p.59 / Chapter 2.10.12 --- 2.11.12Detection with streptavidin-dye conjugate --- p.59 / Chapter 2.10.13 --- Bioarray scanning and analysis --- p.61 / Chapter Chapter 3: --- Results --- p.62 / Chapter 3.1 --- Morphological effects of Lovastatin on human glioblastoma cells --- p.62 / Chapter 3.2 --- Anti-proliferation effects on glioblastoma cell lines --- p.64 / Chapter 3.3 --- Lovastatin-induced caspase3 and 9 activation in human glioblastoma cell lines --- p.69 / Chapter 3.4 --- Cell cycle determination by PI staining --- p.77 / Chapter 3.5 --- Quantification of apoptotic cell death by annexin V and propidium iodide staining --- p.79 / Chapter 3.6 --- Microarray analysis of lovastatin-modulated gene expression profiles --- p.82 / Chapter 3.7 --- Synergistic effects induced by lovastatin and Tumor Necrosis Factor related apoptosis-inducing Ligand (TRAIL) --- p.87 / Chapter 3.7.1 --- M059J and M059K glioblastoma cells was resistant to TRAIL attack --- p.87 / Chapter 3.7.2 --- Synergistic cell death was induced by lovastatin and TRAIL --- p.87 / Chapter 3.7.3 --- A combination of TRAIL and lovastatin induces synergistic apoptosis in glioblastoma cells --- p.93 / Chapter 3.7.4 --- DNA fragmentation on glioblastoma cells --- p.98 / Chapter 3.7.5 --- Four TRAIL receptors mRNA expression profiles on glioblastoma cells --- p.102 / Chapter Chapter 4 --- Discussion --- p.105 / Chapter 4.1 --- Lovastatin exhibited anti-proliferation effects in human glioblastoma cells --- p.107 / Chapter 4.2 --- Lovastatin activated caspase 3 and caspase 9 in human glioblastoma cells --- p.108 / Chapter 4.3 --- Gene expression profile modulated by Lovastatin in human glioblastoma cells --- p.110 / Chapter 4.4 --- Lovastatin-sensitized TRAIL-induced apoptosis in human glioblastoma cells --- p.117 / Chapter Chapter Five: --- Conclusion and Future perspective --- p.121 / References --- p.122 / Appendix --- p.150

Changes in gene expression programs play a central role in cancer. Chromosomal aberrations such as deletions, duplications and translocations of DNA segments can lead to highly significant positive correlations of gene expression levels of neighboring genes. This should be utilized to improve the analysis of tumor expression profiles. Here, we develop a novel model class of autoregressive higher-order Hidden Markov Models (HMMs) that carefully exploit local data-dependent chromosomal dependencies to improve the identification of differentially expressed genes in tumor. Autoregressive higher-order HMMs overcome generally existing limitations of standard first-order HMMs in the modeling of dependencies between genes in close chromosomal proximity by the simultaneous usage of higher-order state-transitions and autoregressive emissions as novel model features. We apply autoregressive higher-order HMMs to the analysis of breast cancer and glioma gene expression data and perform in-depth model evaluation studies. We find that autoregressive higher-order HMMs clearly improve the identification of overexpressed genes with underlying gene copy number duplications in breast cancer in comparison to mixture models, standard first- and higher-order HMMs, and other related methods. The performance benefit is attributed to the simultaneous usage of higher-order state-transitions in combination with autoregressive emissions. This benefit could not be reached by using each of these two features independently. We also find that autoregressive higher-order HMMs are better able to identify differentially expressed genes in tumors independent of the underlying gene copy number status in comparison to the majority of related methods. This is further supported by the identification of well-known and of previously unreported hotspots of differential expression in glioblastomas demonstrating the efficacy of autoregressive higher-order HMMs for the analysis of individual tumor expression profiles. Moreover, we reveal interesting novel details of systematic alterations of gene expression levels in known cancer signaling pathways distinguishing oligodendrogliomas, astrocytomas and glioblastomas.

Le glioblastome multiforme (GBM) est une tumeur de grade IV représentant 49% de toutes les tumeurs cérébrales. Malgré des modalités de traitement agressives (radiothérapie, chimiothérapie et résection chirurgicale), le pronostic est mauvais avec une survie globale médiane de 12 à 14 mois. Les aractéristiques issues de la neuro imagerie des GBM peuvent fournir de nouvelles opportunités pour la classification, le pronostic et le développement de nouvelles thérapies ciblées pour faire progresser la pratique clinique. Cette thèse se concentre sur le développement de modèles pronostiques exploitant des caractéristiques de radiomique extraites des images multimodales IRM (T1 pré- et post-contraste, T2 et FLAIR). Le contexte méthodologique proposé consiste à i) recaler tous les volumes multimodaux IRM disponibles et en segmenter un volume tumoral unique, ii) extraire des caractéristiques radiomiques et iii) construire et valider les modèles pronostiques par l’utilisation d’algorithmes d’apprentissage automatique exploitant des cohortes cliniques multicentriques de patients. Le coeur des méthodes développées est fondé sur l’extraction de radiomiques (incluant des paramètres d’intensité, de forme et de textures) pour construire des modèles pronostiques à l’aide de deux algorithmes d’apprentissage, les machines à vecteurs de support (support vector machines, SVM) et les forêts aléatoires (random forest, RF), comparées dans leur capacité à sélectionner et combiner les caractéristiques optimales. Les bénéfices et l’impact de plusieurs étapes de pré-traitement des images IRM (re-échantillonnage spatial des voxels, normalisation, segmentation et discrétisation des intensités) pour une extraction de métriques fiables ont été évalués. De plus les caractéristiques radiomiques ont été standardisées en participant à l’initiative internationale de standardisation multicentrique des radiomiques. La précision obtenue sur le jeu de test indépendant avec les deux algorithmes d’apprentissage SVM et RF, en fonction des modalités utilisées et du nombre de caractéristiques combinées atteignait 77 à 83% en exploitant toutes les radiomiques disponibles sans prendre en compte leur fiabilité intrinsèque, et 77 à 87% en n’utilisant que les métriques identifiées comme fiables.Dans cette thèse, un contexte méthodologique a été proposé, développé et validé, qui permet la construction de modèles pronostiques dans le cadre des GBM et de l’imagerie multimodale IRM exploitée par des algorithmes d’apprentissage automatique. Les travaux futurs pourront s’intéresser à l’ajout à ces modèles des informations contextuelles et génétiques. D’un point de vue algorithmique, l’exploitation de nouvelles techniques d’apprentissage profond est aussi prometteuse. / Glioblastoma multiforme (GBM) is a WHO grade IV tumor that represents 49% of ail brain tumours. Despite aggressive treatment modalities (radiotherapy, chemotherapy and surgical resections) the prognosis is poor, as médian overall survival (OS) is 12-14 months. GBM’s neuroimaging (non-invasive) features can provide opportunities for subclassification, prognostication, and the development of targeted therapies that could advance the clinical practice. This thesis focuses on developing a prognostic model based on multimodal MRI-derived (Tl pre- and post-contrast, T2 and FLAIR) radiomics in GBM. The proposed methodological framework consists in i) registering the available 3D multimodal MR images andsegmenting the tumor volume, ii) extracting radiomics iii) building and validating a prognostic model using machine learning algorithms applied to multicentric clinical cohorts of patients. The core component of the framework rely on extracting radiomics (including intensity, shape and textural metrics) and building prognostic models using two different machine learning algorithms (Support Vector Machine (SVM) and Random Forest (RF)) that were compared by selecting, ranking and combining optimal features. The potential benefits and respective impact of several MRI pre-processing steps (spatial resampling of the voxels, intensities quantization and normalization, segmentation) for reliable extraction of radiomics was thoroughly assessed. Moreover, the standardization of the radiomics features among methodological teams was done by contributing to “Multicentre Initiative for Standardisation of Radiomics”. The accuracy obtained on the independent test dataset using SVM and RF reached upto 83%- 77% when combining ail available features and upto 87%-77% when using only reliable features previously identified as robust, depending on number of features and modality. In this thesis, I developed a framework for developing a compréhensive prognostic model for patients with GBM from multimodal MRI-derived “radiomics and machine learning”. The future work will consists in building a unified prognostic model exploiting other contextual data such as genomics. In case of new algorithm development we look forward to develop the Ensemble models and deep learning-based techniques.

Changes in gene expression programs play a central role in cancer. Chromosomal aberrations such as deletions, duplications and translocations of DNA segments can lead to highly significant positive correlations of gene expression levels of neighboring genes. This should be utilized to improve the analysis of tumor expression profiles. Here, we develop a novel model class of autoregressive higher-order Hidden Markov Models (HMMs) that carefully exploit local data-dependent chromosomal dependencies to improve the identification of differentially expressed genes in tumor. Autoregressive higher-order HMMs overcome generally existing limitations of standard first-order HMMs in the modeling of dependencies between genes in close chromosomal proximity by the simultaneous usage of higher-order state-transitions and autoregressive emissions as novel model features. We apply autoregressive higher-order HMMs to the analysis of breast cancer and glioma gene expression data and perform in-depth model evaluation studies. We find that autoregressive higher-order HMMs clearly improve the identification of overexpressed genes with underlying gene copy number duplications in breast cancer in comparison to mixture models, standard first- and higher-order HMMs, and other related methods. The performance benefit is attributed to the simultaneous usage of higher-order state-transitions in combination with autoregressive emissions. This benefit could not be reached by using each of these two features independently. We also find that autoregressive higher-order HMMs are better able to identify differentially expressed genes in tumors independent of the underlying gene copy number status in comparison to the majority of related methods. This is further supported by the identification of well-known and of previously unreported hotspots of differential expression in glioblastomas demonstrating the efficacy of autoregressive higher-order HMMs for the analysis of individual tumor expression profiles. Moreover, we reveal interesting novel details of systematic alterations of gene expression levels in known cancer signaling pathways distinguishing oligodendrogliomas, astrocytomas and glioblastomas.

Glioblastoma multiforme (GBM) is the most common and aggressive primary brain tumor. The current standard-of-care treatment including surgery, radiation and temozolomide (TMZ) chemotherapy does not prolong the survival satisfactorily. Here we have tested the feasibility, efficacy and safety of a potential gene therapy approach using AAV as gene delivery vehicle for treatment of GBM. Interferon-beta (IFNβ) is a cytokine molecule also having pleiotropic anticancerous properties. Previously it has been shown by our group that AAV mediated local (intracranial) gene delivery of human IFNβ (hIFNβ) could be an effective treatment for non-invasive human glioblastoma (U87) in orthotopic xenograft mouse model.But as one of the major challenges to treat GBM effectively in clinics is its highly invasive property, in the current study we first sought to test the efficacy of our therapeutic model in a highly invasive human GBM (GBM8) xenograft mouse model. One major limitation of using the xenograft mouse model is that these mice are immune-compromised. Moreover, as IFNβ does not interact with cross-species receptors, the influence of immune systems on GBM remains largely untested. Therefore to test the therapeutic approach in an immune-competent mouse model, we next treated a syngeneic mouse GBM model (GL261) in an immune-competent mouse (C57B6) with the gene encoding the species-matched IFNβ (mIFNβ). We also tested if combination of this IFNβ gene therapy with the current standard chemotherapeutic drug (TMZ) is more effective than any one of the therapeutic modes alone. Finally, we tested the long term safety of the AAV-mIFNβ local gene therapy in healthy C57B6 mice. Next, we hypothesized that global genetic engineering of brain cells expressing secretory therapeutic protein like hIFNβ could be more beneficial for treatment of invasive, migratory and distal multifocal GBM. We tested this hypothesis using systemic delivery of AAV9 vectors encoding hIFNβ gene for treatment of GBM8 tumor in nude mice. Using in vivo bioluminescence imaging of tumor associated firefly luciferase activity, long term survival assay and histological analysis of the brains we have shown that local treatment of AAV-hIFNβ for highly invasive human GBM8 is therapeutically beneficial at an early growth phase of tumor. However, systemic delivery route treatment is far superior for treating multifocal distal GBM8 tumors. Nonetheless, for both delivery routes, treatment efficacy is significantly reduced when treated at a later growth phase of the tumor. In syngeneic GL261 tumor model study, we show that local AAV-mIFNβ gene therapy alone or in combination with TMZ treatment can provide significant survival benefit over control or only TMZ treatment, respectively. However, the animals eventually succumb to the tumor. Safety study in the healthy animals shows significant body weight loss in some treatment groups, whereas one group shows long term survival without any weight loss or any noticeable changes in the external appearances. However, histological analysis indicates marked demyelinating neurotoxic effects upon long term exposures to mIFNβ over-expressions in brain. Overall, we conclude from this study that AAV-IFNβ gene therapy has great therapeutic potential for GBM treatment in future, but the therapeutic window is small and long term continuous expression could have severe deleterious effects on health.

<p>Herein, the development, optimization, and evaluation of quantitative techniques are presented for dynamic PET studies in cancer imaging applications. Dynamic PET image analysis techniques are first applied to 18F-fluoroethyltyrosine (FET) PET imaging of glioma and brain metastasis patients. In a second application, dynamic PET image analysis techniques are applied to 68Ga-PSMA-11 PET imaging for primary prostate cancer patients. Overall, the application of dynamic PET imaging techniques supports improved clinical outcomes and enhanced clinician confidence for treatment modifications. </p>

Un modèle de croissance et de réponse à la radiothérapie pour le glioblastome multiforme (GBM) basé le formalisme du modèle de prolifération-invasion (PI) et du modèle linéaire-quadratique a été développé et implémenté. La géométrie spécifique au patient est considérée en modélisant, d'une part, les voies d'invasion possibles des GBM avec l'imagerie du tenseur de diffusion (DTI) et, d'autre part, les barrières à la propagation à partir des images anatomiques disponibles. La distribution de dose réelle reçue par un patient donné est appliquée telle quelle dans les simulations, en respectant l'horaire de traitement. Les paramètres libres du modèle (taux de prolifération, coefficient de diffusion, paramètres radiobiologiques) sont choisis aléatoirement à partir de distributions de valeurs plausibles. Un total de 400 ensembles de valeurs pour les paramètres libres sont ainsi choisis pour tous les patients, et une simulation de la croissance et de la réponse au traitement est effectuée pour chaque patient et chaque ensemble de paramètres. Un critère de récidive est appliqué sur les résultats de chaque simulation pour identifier un lieu probable de récidive (SPR). La superposition de tous les SPR obtenus pour un patient donné permet de définir la probabilité d'occurrence (OP). Il est démontré qu'il existe des valeurs de OP élevées pour tous les patients, impliquant que les résultats du modèle PI ne sont pas très sensibles aux valeurs des paramètres utilisés. Il est également démontré comment le formalisme développé dans cet ouvrage pourrait permettre de définir un volume cible personnalisé pour les traitements de radiothérapie du GBM. / We have developed and implemented a model of growth and response to radiotherapy for glioblastoma multiforme (GBM) based on the proliferation-invasion (PI) formalism and linear-quadratic model. We take into account patient-specific geometry to model the possible invasion pathways of GBM with diffusion tensor imaging (DTI) and the barriers to dispersal from anatomical images available. The actual dose distribution received by a given patient is applied as such in the simulation, respecting the treatment schedule. The free parameters in the model (proliferation rate, diffusion coefficient, radiobiological parameters) are randomly chosen from a distribution of plausible values. A total of 400 sets of values for the free parameters are thus chosen for all patients, and a simulation of the growth and the response to treatment is performed for each patient and each set of parameters. A failure criterion is applied to the results of each simulation to identify a site of potential recurrence (SPR). The superposition of all SPR obtained for a given patient defines the occurrence probability (OP). We show that high OP values exist for all patients and conclude that the PI model results are not very sensitive to the values of the parameters used. Finally, we show how the formalism developed in this work could help to define a custom target volume for radiation treatment of GBM.