LOW REACTIVE OXYGEN SPECIES AND HIGH GLYCOLYSIS IN GLIOBLASTOMA STEM CELLS:MECHANISMS AND THERAPEUTIC IMPLICATIONS

Glioblastoma multiforme (GBM) is the most common and aggressive primary brain tumor with poor prognosis due in part to drug resistance and high incidence of tumor recurrence. The drug resistant and cancer recurrence phenotype may be ascribed to the presence of glioblastoma stem cells (GSCs), which seem to reside in special stem-cell niches in vivo and require special culture conditions including certain growth factors and serum-free medium to maintain their stemness in vitro. Exposure of GSCs to fetal bovine serum (FBS) can cause their differentiation, the underlying mechanism of which remains unknown. Reactive oxygen species (ROS) play an important role in normal stem cell differentiation, but their role in affecting cancer stem cell fate remains unclear. Whether the metabolic characteristics of GSCs are different from other glioblastoma cells and can be targeted are also unknown. In this study, we used several stem-like glioblastoma cell lines derived from clinical tissues by typical neurosphere culture system or orthotopic xenografts, and showed that addition of fetal bovine serum to the medium induced an increase of ROS, leading to aberrant differentiation and decreases of stem cell markers such as CD133. We found that exposure of GSCs to serum induced their differentiation through activation of mitochondrial respiration, leading to an increase in superoxide (O2-) generation and a profound ROS stress response manifested by upregulation of oxidative stress response pathway. This increase in mitochondrial ROS led to a down-regulation of molecules including SOX2, and Olig2, and Notch1 that are important for stem cell function and an upregulation of mitochondrial superoxide dismutase SOD2 that converts O2- to H2O2. Neutralization of ROS by antioxidant N-acetyl-cysteine in the serum-treated GSCs suppressed the increase of superoxide and partially rescued the expression of SOX2, Olig2, and Notch1, and prevented the serum-induced differentiation phenotype. Additionally, GSCs showed high dependence on glycolysis for energy production. The combination of a glycolytic inhibitor 3-BrOP and a chemotherapeutic agent BCNU depleted cellular ATP and inhibited the repair of BCNU-induced DNA damage, achieving strikingly synergistic killing effects in drug resistant GSCs. This study uncovers the metabolic properties of glioblastoma stem cells and suggests that mitochondrial function and cellular redox status may profoundly affect the fates of glioblastoma stem cells via a ROS-mediated mechanism, and that the active glycolytic metabolism in cancer stem cells may provide a biochemical basis for developing novel therapeutic strategies to effectively eliminate GSCs.

Potential REST Targets in Neural Stem Cells and Glioblastoma-Derived Stem Cells

Glioblastoma multiforme (GBM) tumors are the most common malignant primary brain tumors in adults. The current theory is that these tumors are caused by self-renewing glioblastoma-derived stem cells (GSCs). At the current time, the mechanisms that regulate self-renewal and other oncogenic properties of GSCs remain unknown. Recently, we found transcriptional repressor REST maintains self-renewal in neural stem cells (NSCs) and in GSCs. REST also regulates other oncogenic properties, such as apoptosis, invasion and proliferation. However, the mechanisms by which REST regulates these oncogenic properties are unknown. In an attempt to determine these mechanisms, we performed loss and gain-of-function experiments and genome-wide mRNA expression analysis in GSCs, and we were able to identify REST-regulated genes in GSCs. This was accomplished, after screening concordantly regulated genes in NSCs and GSCs, utilizing two RE1 databases, and setting two-fold expression as filters on the resulting genes. These results received further validation by qRT-PCR. Ingenuity Pathway Analysis (IPA) analysis further revealed the top REST target genes in GSCs were downstream targets of REST and/or involved in other cancers in other cell lines. IPA also revealed that many of the differentially-regulated genes identified in this study are involved in oncogenic properties seen in GBM, and which we believe are related to REST expression.

αvβ8 Integrin Interacts with RhoGDI1 to Regulate Rac1 and Cdc42 Activation and Drive Glioblastoma Cell Invasion

As an interface between the circulatory and central nervous systems, the neurovascular unit is vital to the development and survival of tumors. The malignant brain cancer glioblastoma multiforme (GBM) displays invasive growth behaviors that are major impediments to surgical resection and targeted therapies. Adhesion and signaling pathways that drive GBM cell invasion remain largely uncharacterized. Here we have utilized human GBM cell lines, primary patient samples, and pre-clinical mouse models to demonstrate that integrin αvβ8 is a major driver of GBM cell invasion. β8 integrin is overexpressed in many human GBM cells, with higher integrin expression correlating with increased invasion and diminished patient survival. Silencing β8 integrin in human GBM cells leads to impaired tumor cell invasion due to hyperactivation of the Rho GTPases Rac1 and Cdc42. β8 integrin associates with Rho GDP Dissociation Inhibitor 1 (RhoGDI1), an intracellular signaling effector that sequesters Rho GTPases in their inactive GDP-bound states. Silencing RhoGDI1 expression or uncoupling αvβ8 integrin-RhoGDI1 protein interactions blocks GBM cell invasion due to Rho GTPase hyperactivation. These data reveal for the first time that αvβ8 integrin, via interactions with RhoGDI1, suppresses activation of Rho proteins to promote GBM cell invasiveness. Hence, targeting the αvβ8 integrin-RhoGDI1 signaling axis may be an effective strategy for blocking GBM cell invasion.

YKL -40, a vascular -associated serum marker for glioblastoma, correlates with chronic activation of AKT

YKL-40 is a secreted glycoprotein that has been reported to be expressed in pathologic conditions of extracellular matrix degradation and angiogenesis, such as rheumatoid arthritis, severe osteoarthritis, primary colorectal cancer, metastatic breast cancer, and recurrent ovarian cancer (Dehn, Hogdall et al. 2003). ^ We have identified YKL-40 as a serum marker for glioblastoma multiforme (GBM) using microarray analysis from samples of GBM. We compared the gene expression profile of 19 gliomas to pooled normal brain tissue using the Incyte 10,000 gene expression array. The most differentially expressed gene in this analysis was YKL-40; it was detected in GBM samples with a range of 3 to 62-fold elevation over normal brain. Western blot analysis of glioma samples for YKL-40 protein levels revealed substantial elevation in approximately 65% of GBMs, and undetectable levels in lower-grade gliomas and normal brain tissue. ELISA analysis on serum samples of glioma patients showed that YKL-40 levels were substantially elevated in many of the GBM patients. Statistical analysis indicated that in patients with glioma, serum YKL-40 levels correlate with tumor grade and potentially tumor burden in GBM. ^ Furthermore, we found that YKL-40 expression by in-situ hybridization on a brain tumor tissue array was limited to GBM's and gliosarcomas (GSA), and that YKL-40 expression was specific to the GBM component of GSA. Additional in-situ hybridization analysis, found it to be regionally associated with tumor vasculature as well as activated AKT expression in both human and mouse GBM's. Correlation of elevated YKL-40 with phospho-AKT was confirmed by Western blot analysis on a series of glioblastoma tumors, and inhibition of PI3 Kinase signaling by addition of LY294002 also decreased secretion of YKL-40 over a 7-day period in U87 glioblastoma cell tine. Lastly, YKL-40 expression was induced in response to serum starvation and altered by interaction with specific extracellular matrix (ECM) modules. In summary, we have identified the first accurate serum marker for high-grade gliomas. Furthermore, our findings indicate that YKL-40 is a highly expressed vascular-related glycoprotein in human GBM tissue and that it is affected by the AKT signaling pathway and interaction with components of brain ECM proteins. ^

Suppression of glioblastoma angiogenicity and tumorigenicity by inhibition of endogenous expression of vascular endothelial growth factor.

The development of new capillary networks from the normal microvasculature of the host appears to be required for growth of solid tumors. Tumor cells influence this process by producing both inhibitors and positive effectors of angiogenesis. Among the latter, the vascular endothelial growth factor (VEGF) has assumed prime candidacy as a major positive physiological effector. Here, we have directly tested this hypothesis in the brain tumor, glioblastoma multiforme, one of the most highly vascularized human cancers. We introduced an antisense VEGF expression construct into glioblastoma cells and found that (i) VEGF mRNA and protein levels were markedly reduced, (ii) the modified cells did not secrete sufficient factors so as to be chemoattractive for primary human microvascular endothelial cells, (iii) the modified cells were not able to sustain tumor growth in immunodeficient animals, and (iv) the density of in vivo blood vessel formation was reduced in direct relation to the reduction of VEGF secretion and tumor formation. Moreover, revertant cells that recovered the ability to secrete VEGF regained each of these tumorigenic properties. These results suggest that VEGF plays a major angiogenic role in glioblastoma.

Reverse Engineering of Modified Genes by Bayesian Network Analysis Defines Molecular Determinants Critical to the Development of Glioblastoma

In this study we have identified key genes that are critical in development of astrocytic tumors. Meta-analysis of microarray studies which compared normal tissue to astrocytoma revealed a set of 646 differentially expressed genes in the majority of astrocytoma. Reverse engineering of these 646 genes using Bayesian network analysis produced a gene network for each grade of astrocytoma (Grade I–IV), and ‘key genes’ within each grade were identified. Genes found to be most influential to development of the highest grade of astrocytoma, Glioblastoma multiforme were: COL4A1, EGFR, BTF3, MPP2, RAB31, CDK4, CD99, ANXA2, TOP2A, and SERBP1. All of these genes were up-regulated, except MPP2 (down regulated). These 10 genes were able to predict tumor status with 96–100% confidence when using logistic regression, cross validation, and the support vector machine analysis. Markov genes interact with NFkβ, ERK, MAPK, VEGF, growth hormone and collagen to produce a network whose top biological functions are cancer, neurological disease, and cellular movement. Three of the 10 genes - EGFR, COL4A1, and CDK4, in particular, seemed to be potential ‘hubs of activity’. Modified expression of these 10 Markov Blanket genes increases lifetime risk of developing glioblastoma compared to the normal population. The glioblastoma risk estimates were dramatically increased with joint effects of 4 or more than 4 Markov Blanket genes. Joint interaction effects of 4, 5, 6, 7, 8, 9 or 10 Markov Blanket genes produced 9, 13, 20.9, 26.7, 52.8, 53.2, 78.1 or 85.9%, respectively, increase in lifetime risk of developing glioblastoma compared to normal population. In summary, it appears that modified expression of several ‘key genes’ may be required for the development of glioblastoma. Further studies are needed to validate these ‘key genes’ as useful tools for early detection and novel therapeutic options for these tumors.

Toward an effective strategy in glioblastoma treatment. Part II: RNA interference as a promising way to sensitize glioblastomas to temozolomide.

International audience

Toward an effective strategy in glioblastoma treatment. Part I: resistance mechanisms and strategies to overcome resistance of glioblastoma to temozolomide.

International audience

RARβ gene methylation is a candidate for primary glioblastoma treatment planning.

Background: We screened RARβ methylation in primary glioblastoma multiforme (GBM) and the results were evaluated based on the clinical data and treatment type. Objective: The objective of this study was to find new areas for the usage of MS-HRM applications in the determination of methylation levels in primary GBM samples and it shows the association of RARβ methylation with the clinical outcome. Methods: In our study, tumor samples were collected during surgical resection by the Department of Neurosurgery. The clinical and radiologic data was carefully reviewed, compared, and evaluated with the histological results. The methylation status of RARβ was determined by using MS-HRM. Results: RARβ gene methylation was detected in 24 out of 40 cases (60%), with different quantitative methylation levels. The mean survival time was 19 months form ethylated cases and 15 months for the non-methylated cases. The survival time of the patients who received treatment was 25 months and the survival time of the patients who received radiotherapy alone or where no treatment protocol applied was 15-20 months. Therefore, a significant difference in survival rates has been observed (P<0.05). This study indicates a potential prognostic value for GBM treatment planning. Conclusion: Our study is the first study to investigate RARβ methylation in primary GBMs. We conclude that the RARβ gene could be a new prognostic and predictive candidate marker to designate the treatment protocol for primary GBMs. Keywords:

The mechanical rigidity of the extracellular matrix regulates the structure, motility, and proliferation of glioma cells

Glioblastoma multiforme (GBM) is a malignant astrocytoma of the central nervous system associated with a median survival time of 15 months, even with aggressive therapy. This rapid progression is due in part to diffuse infiltration of single tumor cells into the brain parenchyma, which is thought to involve aberrant interactions between tumor cells and the extracellular matrix (ECM). Here, we test the hypothesis that mechanical cues from the ECM contribute to key tumor cell properties relevant to invasion. We cultured a series of glioma cell lines (U373-MG, U87-MG, U251-MG, SNB19, C6) on fibronectin-coated polymeric ECM substrates of defined mechanical rigidity and investigated the role of ECM rigidity in regulating tumor cell structure, migration, and proliferation. On highly rigid ECMs, tumor cells spread extensively, form prominent stress fibers and mature focal adhesions, and migrate rapidly. As ECM rigidity is lowered to values comparable with normal brain tissue, tumor cells appear rounded and fail to productively migrate. Remarkably, cell proliferation is also strongly regulated by ECM rigidity, with cells dividing much more rapidly on rigid than on compliant ECMs. Pharmacologic inhibition of nonmuscle myosin II–based contractility blunts this rigidity-sensitivity and rescues cell motility on highly compliant substrates. Collectively, our results provide support for a novel model in which ECM rigidity provides a transformative, microenvironmental cue that acts through actomyosin contractility to regulate the invasive properties of GBM tumor cells.

Proton magnetic resonance spectroscopy of a boron neutron capture therapy 10B-carrier, L-p-boronophenylalanine-fructose complex

Boron neutron capture therapy (BNCT) is a radiotherapy that has mainly been used to treat malignant brain tumours, melanomas, and head and neck cancer. In BNCT, the patient receives an intravenous infusion of a 10B-carrier, which accumulates in the tumour area. The tumour is irradiated with epithermal or thermal neutrons, which result in a boron neutron capture reaction that generates heavy particles to damage tumour cells. In Finland, boronophenylalanine fructose (BPA-F) is used as the 10B-carrier. Currently, the drifting of boron from blood to tumour as well as the spatial and temporal accumulation of boron in the brain, are not precisely known. Proton magnetic resonance spectroscopy (1H MRS) could be used for selective BPA-F detection and quantification as aromatic protons of BPA resonate in the spectrum region, which is clear of brain metabolite signals. This study, which included both phantom and in vivo studies, examined the validity of 1H MRS as a tool for BPA detection. In the phantom study, BPA quantification was studied at 1.5 and 3.0 T with single voxel 1H MRS, and at 1.5 T with magnetic resonance imaging (MRSI). The detection limit of BPA was determined in phantom conditions at 1.5 T and 3.0 T using single voxel 1H MRS, and at 1.5 T using MRSI. In phantom conditions, BPA quantification accuracy of ± 5% and ± 15% were achieved with single voxel MRS using external or internal (internal water signal) concentration references, respectively. For MRSI, a quantification accuracy of <5% was obtained using an internal concentration reference (creatine). The detection limits of BPA in phantom conditions for the PRESS sequence were 0.7 (3.0 T) and 1.4 mM (1.5 T) mM with 20 × 20 × 20 mm3 single voxel MRS, and 1.0 mM with acquisition-weighted MRSI (nominal voxel volume 10(RL) × 10(AP) × 7.5(SI) mm3), respectively. In the in vivo study, an MRSI or single voxel MRS or both was performed for ten patients (patients 1-10) on the day of BNCT. Three patients had glioblastoma multiforme (GBM), and five patients had a recurrent or progressing GBM or anaplastic astrocytoma gradus III, and two patients had head and neck cancer. For nine patients (patients 1-9), MRS/MRSI was performed 70-140 min after the second irradiation field, and for one patient (patient 10), the MRSI study began 11 min before the end of the BPA-F infusion and ended 6 min after the end of the infusion. In comparison, single voxel MRS was performed before BNCT, for two patients (patients 3 and 9), and for one patient (patient 9), MRSI was performed one month after treatment. For one patient (patient 10), MRSI was performed four days before infusion. Signals from the tumour spectrum aromatic region were detected on the day of BNCT in three patients, indicating that in favourable cases, it is possible to detect BPA in vivo in the patient’s brain after BNCT treatment or at the end of BPA-F infusion. However, because the shape and position of the detected signals did not exactly match the BPA spectrum detected in the in vitro conditions, assignment of BPA is difficult. The opportunity to perform MRS immediately after the end of BPA-F infusion for more patients is necessary to evaluate the suitability of 1H MRS for BPA detection or quantification for treatment planning purposes. However, it could be possible to use MRSI as criteria in selecting patients for BNCT.

Radiotherapy plus nimotuzumab or placebo in the treatment of high grade glioma patients: results from a randomized, double blind trial

Background The prognosis of patients bearing high grade glioma remains dismal. Epidermal Growth Factor Receptor (EGFR) is well validated as a primary contributor of glioma initiation and progression. Nimotuzumab is a humanized monoclonal antibody that recognizes the EGFR extracellular domain and reaches Central Nervous System tumors, in nonclinical and clinical setting. While it has similar activity when compared to other anti-EGFR antibodies, it does not induce skin toxicity or hypomagnesemia. Methods A randomized, double blind, multicentric clinical trial was conducted in high grade glioma patients (41 anaplastic astrocytoma and 29 glioblastoma multiforme) that received radiotherapy plus nimotuzumab or placebo. Treatment and placebo groups were well-balanced for the most important prognostic variables. Patients received 6 weekly doses of 200 mg nimotuzumab or placebo together with irradiation as induction therapy. Maintenance treatment was given for 1 year with subsequent doses administered every 3 weeks. The objectives of this study were to assess the comparative overall survival, progression free survival, response rate, immunogenicity and safety. Results The median cumulative dose was 3200 mg of nimotuzumab given over a median number of 16 doses. The combination of nimotuzumab and RT was well-tolerated. The most prevalent related adverse reactions included nausea, fever, tremors, anorexia and hepatic test alteration. No anti-idiotypic response was detected, confirming the antibody low immunogenicity. The mean and median survival time for subjects treated with nimotuzumab was 31.06 and 17.76 vs. 21.07 and 12.63 months for the control group. Conclusions In this randomized trial, nimotuzumab showed an excellent safety profile and significant survival benefit in combination with irradiation.

Technology for Single Cell Protein Analysis in Immunology and Cancer Prognostics

The first chapter of this thesis deals with automating data gathering for single cell microfluidic tests. The programs developed saved significant amounts of time with no loss in accuracy. The technology from this chapter was applied to experiments in both Chapters 4 and 5.

The second chapter describes the use of statistical learning to prognose if an anti-angiogenic drug (Bevacizumab) would successfully treat a glioblastoma multiforme tumor. This was conducted by first measuring protein levels from 92 blood samples using the DNA-encoded antibody library platform. This allowed the measure of 35 different proteins per sample, with comparable sensitivity to ELISA. Two statistical learning models were developed in order to predict whether the treatment would succeed. The first, logistic regression, predicted with 85% accuracy and an AUC of 0.901 using a five protein panel. These five proteins were statistically significant predictors and gave insight into the mechanism behind anti-angiogenic success/failure. The second model, an ensemble model of logistic regression, kNN, and random forest, predicted with a slightly higher accuracy of 87%.

The third chapter details the development of a photocleavable conjugate that multiplexed cell surface detection in microfluidic devices. The method successfully detected streptavidin on coated beads with 92% positive predictive rate. Furthermore, chambers with 0, 1, 2, and 3+ beads were statistically distinguishable. The method was then used to detect CD3 on Jurkat T cells, yielding a positive predictive rate of 49% and false positive rate of 0%.

The fourth chapter talks about the use of measuring T cell polyfunctionality in order to predict whether a patient will succeed an adoptive T cells transfer therapy. In 15 patients, we measured 10 proteins from individual T cells (~300 cells per patient). The polyfunctional strength index was calculated, which was then correlated with the patient's progress free survival (PFS) time. 52 other parameters measured in the single cell test were correlated with the PFS. No statistical correlator has been determined, however, and more data is necessary to reach a conclusion.

Finally, the fifth chapter talks about the interactions between T cells and how that affects their protein secretion. It was observed that T cells in direct contact selectively enhance their protein secretion, in some cases by over 5 fold. This occurred for Granzyme B, Perforin, CCL4, TNFa, and IFNg. IL- 10 was shown to decrease slightly upon contact. This phenomenon held true for T cells from all patients tested (n=8). Using single cell data, the theoretical protein secretion frequency was calculated for two cells and then compared to the observed rate of secretion for both two cells not in contact, and two cells in contact. In over 90% of cases, the theoretical protein secretion rate matched that of two cells not in contact.

Brain tumor and brain endothelial cells' response to ionizing radiation and phytochemical treatments

Le glioblastome multiforme (GBM) représente la tumeur cérébrale primaire la plus agressive et la plus vascularisée chez l’adulte. La survie médiane après le diagnostic est de moins d’un an en l’absence de traitement. Malheureusement, 90% des patients traités avec de la radiothérapie après la résection chirurgicale d’un GBM développent une récidive tumorale. Récemment, le traitement des GBM avec radiothérapie et témozolomide, un agent reconnu pour ses propriétés antiangiogéniques, a permis de prolonger la survie médiane à 14,6 mois. Des efforts sont déployés pour identifier des substances naturelles capables d’inhiber, de retarder ou de renverser le processus de carcinogenèse. Epigallocatechin-3-gallate (EGCG), un polyphénol retrouvé dans le thé vert, est reconnu pour ses propriétés anticancéreuses et antiangiogéniques. L’EGCG pourrait sensibiliser les cellules tumorales cérébrales et les cellules endothéliales dérivées des tumeurs aux traitements conventionnels. Le chapitre II décrit la première partie de ce projet de doctorat. Nous avons tenté de déterminer si l’EGCG pourrait sensibiliser la réponse des GBM à l’irradiation (IR) et si des marqueurs moléculaires spécifiques sont impliqués. Nous avons documenté que les cellules U-87 étaient relativement radiorésistantes et que Survivin, une protéine inhibitrice de l’apoptose, pourrait être impliquée dans la radiorésistance des GBM. Aussi, nous avons démontré que le pré-traitement des cellules U-87 avec de l’EGCG pourrait annuler l’effet cytoprotecteur d’une surexpression de Survivin et potentialiser l’effet cytoréducteur de l’IR. Au chapitre III, nous avons caractérisé l’impact de l’IR sur la survie de cellules endothéliales microvasculaires cérébrales humaines (HBMEC) et nous avons déterminé si l’EGCG pouvait optimiser cet effet. Bien que les traitements individuels avec l’EGCG et l’IR diminuaient la survie des HBMEC, le traitement combiné diminuait de façon synergique la survie cellulaire. Nous avons documenté que le traitement combiné augmentait la mort cellulaire, plus spécifiquement la nécrose. Au chapitre IV, nous avons investigué l’impact de l’IR sur les fonctions angiogéniques des HBMEC résistantes à l’IR, notamment la prolifération cellulaire, la migration cellulaire en présence de facteurs de croissance dérivés des tumeurs cérébrales, et la capacité de tubulogenèse. La voie de signalisation des Rho a aussi été étudiée en relation avec les propriétés angiogéniques des HBMEC radiorésistantes. Nos données suggèrent que l’IR altère significativement les propriétés angiogéniques des HBMEC. La réponse aux facteurs importants pour la croissance tumorale et l’angiogenèse ainsi que la tubulogenèse sont atténuées dans ces cellules. En conclusion, ce projet de doctorat confirme les propriétés cytoréductrices de l’IR sur les gliomes malins et propose un nouveau mécanisme pour expliquer la radiorésistance des GBM. Ce projet documente pour la première fois l’effet cytotoxique de l’IR sur les HBMEC. Aussi, ce projet reconnaît l’existence de HBMEC radiorésistantes et caractérise leurs fonctions angiogéniques altérées. La combinaison de molécules naturelles anticancéreuses et antiangiogéniques telles que l’EGCG avec de la radiothérapie pourrait améliorer l’effet de l’IR sur les cellules tumorales et sur les cellules endothéliales associées, possiblement en augmentant la mort cellulaire. Cette thèse supporte l’intégration de nutriments avec propriétés anticancéreuses et antiangiogéniques dans le traitement des gliomes malins pour sensibiliser les cellules tumorales et endothéliales aux traitements conventionnels.

Modélisation radiobiologique pour la planification des traitements en radiothérapie à partir de données d’imagerie spécifiques aux patients

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.