Search

We previously showed that the farnesyl transferase inihibitor, Tipifarnib induced vascularization normalization, oxygenation and radiosensitization in a pre-clinical glioblastoma (GBM) model. The aim of this study was to assess by dynamic-susceptibility-contrast MRI (DSC-MRI) the effect of radiotherapy (RT) and Tipifarnib combination on tumor perfusion in GBM patients. Eighteen patients with newly diagnosed GBM, enrolled in a phase I-II clinical trial associating RT with Tipifarnib, underwent anatomical MR imaging and DSC-MRI before (M0) and two months after treatment (M2). Anatomic volumes of interest (VOIs) were delineated according to contrast-enhanced and hyper-intense signal areas on T1-Gd and T2 images, respectively. Perfusion variations between M0 and M2 were assessed with median relative cerebral blood volume (rCBV) inside these VOIs. Another voxel by voxel analysis of CBV values classified 405,117 tumor voxels into High_, Normal_ and Low_CBVTUMOR according to the distribution of CBV in the contralateral normal tissue. These three categories of CBVTUMOR voxels were color-coded over anatomical MRI. Variations of median rCBV were significantly different for two groups of patients (P 0.013): rCBV decreased when initial rCBV was ≥ 1.0 (Group_rCBV_M0 1) and rCBV increased when initial rCBV was 1.0 (Group_rCBV_M0 1). Mapping of color-coded voxels provided additional spatial and quantitative information about tumor perfusion: Group_rCBV_M0 1 presented a significant decrease of High_CBVTUMOR volume (P = 0.015) simultaneously with a significant increase of Normal_CBVTUMOR volume (P = 0.009) after treatment. Group_rCBV_M0 1 presented a decrease of Low_CBVTUMOR volume with an increase of Normal_ and High_CBV TUMOR volume after treatment. Pre and post-treatment CBV measurements with DSC-MRI characterized tumor perfusion evolution in GBM patients treated with RT combined to Tipifarnib; showing variations in favour of tumor perfusion normalization in agreement with our pre-clinical results of vascular normalization.

Purpose Because lactate accumulation is considered a surrogate for hypoxia and tumor radiation resistance, we studied the spatial distribution of the lactate-to- N -acetyl-aspartate ratio (LNR) before radiation therapy (RT) with 3D proton magnetic resonance spectroscopic imaging (3D- 1 H-MRSI) and assessed its impact on local tumor control in glioblastoma (GBM). Methods and Materials Fourteen patients with newly diagnosed GBM included in a phase 2 chemoradiation therapy trial constituted our database. Magnetic resonance imaging (MRI) and MRSI data before RT were evaluated and correlated to MRI data at relapse. The optimal threshold for tumor-associated LNR was determined with receiver-operating-characteristic (ROC) curve analysis of the pre-RT LNR values and MRI characteristics of the tumor. This threshold was used to segment pre-RT normalized LNR maps. Two spatial analyses were performed: ( 1 ) a pre-RT volumetric comparison of abnormal LNR areas with regions of MRI-defined lesions and a choline (Cho)-to- N -acetyl-aspartate (NAA) ratio ≥2 (CNR2); and ( 2 ) a voxel-by-voxel spatial analysis of 4,186,185 voxels with the intention of evaluating whether pre-RT abnormal LNR areas were predictive of the site of local recurrence. Results A LNR of ≥0.4 (LNR-0.4) discriminated between tumor-associated and normal LNR values with 88.8% sensitivity and 97.6% specificity. LNR-0.4 voxels were spatially different from those of MRI-defined lesions, representing 44% of contrast enhancement, 64% of central necrosis, and 26% of fluid-attenuated inversion recovery (FLAIR) abnormality volumes before RT. They extended beyond the overlap with CNR2 for most patients (median: 20 cm 3 ; range: 6-49 cm 3 ). LNR-0.4 voxels were significantly predictive of local recurrence, regarded as contrast enhancement at relapse: 71% of voxels with a LNR-0.4 before RT were contrast enhanced at relapse versus 10% of voxels with a normal LNR ( P Conclusions Pre-RT LNR-0.4 in GBM indicates tumor areas that are likely to relapse. Further investigations are needed to confirm lactate imaging as a tool to define additional biological target volumes for dose painting.

Background To integrate 3D MR spectroscopy imaging (MRSI) in the treatment planning system (TPS) for glioblastoma dose painting to guide simultaneous integrated boost (SIB) in intensity-modulated radiation therapy (IMRT). Methods For sixteen glioblastoma patients, we have simulated three types of dosimetry plans, one conventional plan of 60-Gy in 3D conformational radiotherapy (3D-CRT), one 60-Gy plan in IMRT and one 72-Gy plan in SIB-IMRT. All sixteen MRSI metabolic maps were integrated into TPS, using normalization with color-space conversion and threshold-based segmentation. The fusion between the metabolic maps and the planning CT scans were assessed. Dosimetry comparisons were performed between the different plans of 60-Gy 3D-CRT, 60-Gy IMRT and 72-Gy SIB-IMRT, the last plan was targeted on MRSI abnormalities and contrast enhancement (CE). Results Fusion assessment was performed for 160 transformations. It resulted in maximum differences p Conclusions Delivering standard doses to conventional target and higher doses to new target volumes characterized by MRSI and CE is now possible and does not increase dose to organs at risk. MRSI and CE abnormalities are now integrated for glioblastoma SIB-IMRT, concomitant with temozolomide, in an ongoing multi-institutional phase-III clinical trial. Our method of MR spectroscopy maps integration to TPS is robust and reliable; integration to neuronavigation systems with this method could also improve glioblastoma resection or guide biopsies.

X. Franceries1,2,4, N. Chauveau1,2,*, A. Sors3, M. Masquere4 and P. Celsis1,2 1Inserm, Imagerie Cerebrale et Handicaps Neurologiques UMR 825, Toulouse 2Universite de Toulouse, UPS, Imagerie Cerebrale et Handicaps Neurologiques UMR 825, CHU Purpan, Place du Dr Baylac, Toulouse Cedex 9 3LU 48 LERISM Laboratoire d’Etudes et de Recherche en Imagerie Spatiale et Medicale, UPS, Toulouse Cedex 4 4Universite de Toulouse, UPS, INPT, LAPLACE (Laboratoire Plasma et Conversion d'Energie), Toulouse Cedex 9 France

OBJECTIVES: To evaluate the changes in regional cerebral blood flow (rCBF) and memory performance in patients with age related cognitive decline (ARCD) who did and did not become demented during a follow up period. METHODS: Twenty four patients with ARCD were recruited from an outpatient memory clinic, of whom 18 were followed up over a mean period of two years. Eighteen patients with mild to moderate probable Alzheimer's disease and 18 aged normal controls were followed up over a mean period of three years. Memory performance and rCBF were evaluated quantitatively at inclusion and during follow up, using single photon emission computed tomography with xenon-133 injection and three subtests of the Wechsler memory scale (logical memory, paired associated learning, and digit span). RESULTS: Patients with ARCD showed decreased rCBF and memory performance at initial evaluation compared with controls. Five of them became demented during the follow up period, with further decline in memory and rCBF. At inclusion, the only feature that distinguished these five patients as a group from the remainder was a pronounced temporoparietal asymmetry. The 13 patients with ARCD who did not become demented still exhibited impaired memory and rCBF at follow up, but without any further decline and no increase in flow asymmetry. CONCLUSIONS: Apart from patients in the preclinical phase of Alzheimer's disease, the ARCD category includes non-demented patients who have brain dysfunction that may represent a distinct clinical entity.

Introduction While conflict between instructions and distractors makes normally reactions slower, right inferior parietal lobule (IPL) damage associated with left spatial neglect leads, in a visuomotor task, to the paradoxical facilitation of rightwards movements in the presence of conflicting leftward response plans (Coulthard et al., 2008). We investigated the effects of the right IPL lesion on the interference between vision and audition in a spatial context. Methods Behavioral data and Event-Related Potentials (ERPs) of 12 healthy adults and 12 patients with a focal right parietal post-stroke lesion, matched in age and educational level, were recorded during an audio-visual task. A pure tone through a stereo system and a white circle on a screen were presented simultaneously on the same side (left or right) in congruent trials and on opposite sides in incongruent trials. Participants had to respond to the auditory or visual target's side according to the instruction, by pressing a response-pad button with their right hand. Whereas all the patients had a left spatial neglect at the acute phase of stroke, they had recovered a normal performance on “paper-and-pencil” tests when included in our study. Results Paradoxically, for right visual target with left auditory distractor, patients had a smaller reaction time “cost” than the control group. For healthy controls, two frontocentral ERPs negative components, N2 and N270, were elicited in all conflict conditions compared to congruent conditions. Whereas the amplitude of N2 and N270 was not different between the two groups for left targets, these components were not generated by the presentation of a right visual or auditory target with a left distractor in patients. Discussion N2 and N270 might be elicited by the detection of two incongruent response plans by the prefrontal cortex. Their suppression for left targets in patients supports the hypothesis that the right IPL might be a crucial site for automatic activation of competing motor plans generated by cross-modal conflicting stimuli. In patients, the behavioral facilitation for right visual targets only may be due to additional effect of visual dominance over audition in spatial tasks.

With an initial formation in theoretical computer sciences with a focus on image processing and analysis, my current research activities deal with image and information processing and analysis for applications in medicine, namely oncology and radiotherapy. More specifically, my research interests are image automatic segmentation and classification for organs and tumors delineation, image denoising and deconvolution for qualitative and quantitative improvement, and more recently, multi observation models for multi modal imaging and information fusion for computer-aided decision making in patients management. These developments are especially considered within the context of the use of Positron Emission Tomography and Computed Tomography (PET/CT) for oncology and radiotherapy applications. My research activities have been and are still carried out within the team "quantitative multi modal imaging for diagnosis and therapy", in the LaTIM INSERM U1101. This framework ensures a multi disciplinary teamwork, in collaboration with radiation oncologists, nuclear medicine physicians, physicists, engineers, mathematicians and computer scientists. As a PhD student, my main contribution to the field has been the development of image segmentation algorithms dedicated to the automated delineation of metabolically active tumor volumes in PET images, with a specific focus on adapting the methodology to specific characteristics of the processed images. As a post-doctoral fellow, I have been further investigating the accuracy, robustness and reproducibility of this methodology within a project for which I had obtained funding from the French research agency (ANR) for two and a half years. I have also been investigating the impact of such methodology and its resulting tumor volumes measurements in various applications such as the dosimetry impact in radiotherapy treatment planning or patient management and therapy assessment in oncology. During the last six years I have also been more and more involved in research developments by several PhD students and post-doctoral fellows, such as PET images fusion for quantitative follow up, realistic simulations and dosimetry evaluation, methodologies for reduction of partial volume effects and noise in emission imaging, and textural features analysis for characterization of tracer uptake heterogeneity within tumors. Overall, I have been acting as co-supervisor of most of the trainees and PhD students of the team under the direction of Research Director D. Visvikis. This responsibility included weekly meetings and discussions with the students, help with designing experiments and analyzing the results, writing of thesis reports and research articles, as well as finding solutions to theoretical problems and technical issues. In the next few years, my research project will be focused on the development of a robust and flexible framework for the modeling and the automatic analysis of multi modality, multi resolution, multi observation and multi temporal images datasets, such as PET/CT, SPECT/CT, PET/MRI, multi MRI or PET tracers imaging, as well as dynamic acquisitions. This development will allow deriving new predictive models and decision tools by fusion of the available multi source information (imaging, genetics and other clinical data), validated in various models such as esophageal, rectal, lung, or head & neck cancers. Additional applications in neurology (for example Alzheimer's disease) might also be explored. This project will be based on previous developments by PhD students that are almost finished with their thesis, and will focus on new PhD students beginning their work under my supervision in October 2011 and early 2012, whose recruitment was possible thanks to grants I contributed to obtain in 2010-2011., Avec une formation initiale en sciences de l'informatique et une spécialisation image, mes activités de recherche actuelles concernent le traitement et l'analyse de l'information et de l'image pour des applications en médecine, plus particulièrement l'oncologie et la radiothérapie. Plus spécifiquement, je m'intéresse à la segmentation et la classification automatique pour la définition des contours d'organes et de tumeurs, au filtrage du bruit et à la déconvolution pour l'amélioration qualitative et quantitative, et plus récemment, aux modèles multi observation pour la prise en compte des images multi modales, et la fusion d'informations pour l'aide à la décision dans la prise en charge des patients. Je poursuis ces thématiques spécifiquement dans le cadre de l'utilisation de l'imagerie TEP/TDM (Tomographie par Emission de Positons et scanner X) en oncologie et radiothérapie. Mes activités de recherche ont pris place dans le contexte de l'équipe " imagerie multi modale quantitative pour le diagnostic et la thérapie " du laboratoire INSERM U650 de traitement de l'information médicale (LaTIM). Ce contexte a garantit un travail l'équipe pluridisciplinaire, en collaboration notamment avec des radiothérapeutes, des médecins nucléaires, des physiciens, des ingénieurs, des mathématiciens et des informaticiens. En tant que doctorant, ma principale contribution a été le développement d'une méthode originale de segmentation d'image adaptée à la définition des volumes fonctionnels des tumeurs sur les images TEP. Lors de mon post-doctorat, j'ai poursuivi la validation de la précision, de la robustesse et de la reproductibilité de cette approche dans le cadre d'un projet ANR pour lequel j'ai reçu un financement de deux ans et demi. J'ai également étudié au cours de ces deux dernières années l'impact d'une telle méthode dans de nombreuses applications, telles que la dosimétrie en planification de traitement en radiothérapie, et la prise en charge des patients en oncologie. Au cours de ces six dernières années, j'ai été de plus en plus impliqué dans des travaux de recherche connexes menés par d'autres doctorants et post-doctorants. Ces travaux incluent la fusion d'images TEP pour le suivi temporel quantitatif, les simulations réalistes et l'évaluation dosimétrique, la caractérisation de l'hétérogénéité intra tumorale des traceurs TEP par analyse de texture, et la réduction des effets de volume partiel et du bruit en imagerie d'émission. J'ai assumé la responsabilité de co-encadrant de plusieurs stagiaires et doctorants de l'équipe sous la direction du directeur de recherche D. Visvikis. Cette responsabilité inclus des réunions hebdomadaires et des discussions régulières avec les étudiants, l'aide à la mise en place des expériences et protocoles de validation, à l'analyse des résultats, la correction de rapports de stage, d'articles et de manuscrits de thèse, ainsi que réfléchir à des solutions aux problèmes tant théoriques que techniques. Je travaille actuellement en tant que chercheur associé au département de recherche en imagerie et radiothérapie de Maastricht (MAASTRO) aux Pays-bas. Au cours des prochaines années, mon projet de recherche sera dédié au développement d'un contexte flexible et robuste permettant la modélisation et l'analyse semi-automatique d'ensemble d'images médicales multi modales, multi résolutions et multi temporelles, telles que TEP/TDM, TEMP/TDM, TEP/IRM, multi IRM ou TEP avec différents traceurs, ainsi que des acquisitions dynamiques. Ce développement permettra de déduire de nouveaux modèles prédictifs et des outils de décision adaptés à diverses applications cliniques tels que les cancers de l'oesophage, rectal, pulmonaire ou ORL, par la fusion de toute l'information disponible (imagerie, génétique, phéntypes et rapports cliniques). Ce projet se construira en partie sur les travaux préliminaires réalisés avec des doctorants venant de soutenir et en passe de terminer leur thèse, et sur les thèses de deux nouvelles doctorantes que j'encadrerai à partir d'octobre 2011 et courant 2012, recrutées sur des financements que j'ai contribué à obtenir en 2010-2011.