Le Cancer, l’une des pathologies les plus fréquentes au sein de la population, possède encore actuellement un taux de morbi-mortalité important tous sexes confondus, et ce malgré les importants progrès diagnostiques et thérapeutiques réalisés. D’un point de vue diagnostique, dans une approche dite de « Biologie de systèmes », en complément de l’étude anatomo- pathologique qui reste la référence, de nouvelles techniques ont été développées pour la caractérisation de profils métaboliques (Métabolomique) d’échantillons tissulaires pathologiques ou non, parmi lesquelles la Spectroscopie RMN HRMAS. Après un bref rappel théorique et avoir dressé le bilan des applications de cette technique en Cancérologie, nous avons exposé les différentes étapes du protocole à mettre en place afin d’envisager son implémentation dans un cadre hospitalier. L’ensemble des résultats présentés permettent d’envisager l’utilisation de cette technique en pratique clinique courante. Il faut néanmoins valider la robustesse des modèles statistiques élaborés et confirmer ces résultats sur de plus grandes cohortes d’échantillons. Des développements technologiques, analytiques et statistiques sont également nécessaires. / Cancer, one of the most frequent pathologies among the population, has still an important morbidity-mortality rate all sex confounded, despite the important diagnostical and therapeutical progresses achieved. From a diagnostical point of view, in a so called “Systems Biology approach”, as a complement of the gold standard histopathological study, some new techniques have been developed for the characterization of metabolic profiles (Metabolomics) of tissular samples pathological or not, among which HRMAS NMR Spectroscopy. After some brief theoretical considerations and after reporting the applications of this technique in Cancerology, we exposed the different steps of the protocol to design in order to consider its implementation in a hospital set up. All the results presented allow considering the use of this technique in a clinical routine. Nevertheless, it is necessary to validate the robustness of the statistical models built and to confirm these results on much larger cohorts of samples. Some technical, analytical and statistical developments are also needed.

La spectroscopie RMN Haute Résolution en Rotation à l’Angle Magique (HRMAS) permet la caractérisation métabolomique tissulaire. Nous avons caractérisé par RMN HRMAS le profil métabolomique des glandes parathyroïdiennes hypersécrétantes. Celui de l’hyperparathyroïdie primaire (HPT1) a été comparé à celui des HPT rénales. Au sein des HPT1, la distinction a pu être faite entre pathologie uni- et multi-glandulaire. Le profil métabolomique du tissu pancréatique sain a été comparé à celui du tissu tumoral. Aussi, les patients longs/courts-survivants ont pu être distingués. La relation entre le phénotype métabolique et la survie des patients a été étudiée. Le profil métabolomique des TIPMP a été caractérisé. Les TIPMP non dégénérées et dégénérées ont été comparées. Le risque de dégénérescence a été corrélé au profil métabolomique. Nos résultats montrent que la spectroscopie RMN HRMAS est une technique prometteuse pour l’étude du profil métabolomique des HPT et des pathologies pancréatiques. / High Resolution Magic Angle Spinning (HRMAS) NMR spectroscopy allows metabolomics of intact tissues. Metabolomics profiling of hyperfunctioning parathyroid glands were characterized were characterized by using HRMAS NMR spectroscopy. Primary hyperparathy- roidism (PHPT) was compared to renal HPT. Among PHPT, we distinguished single gland disease from multiple gland disease. Pancreatic parenchyma and adenocarcinoma were compared. Thus, long-term and short-term survival patients were distinguished. The relationship between the survival of patients and their metabolic phenotype was studied. Metabolomics profiling of IPMN was also examined. IPMN with no degeneration and de- generated IPMN were compared. Finally, the risk of degeneration was correlated with the metabolomics profile. Our results show that HRMAS NMR spectroscopy is a promising technique in view of studying metabolomic profiling of HPT and pancreatic diseases.

Le cancer est la première cause de morbi-mortalité dans le monde. La recherche de biomarqueurs diagnostiques, pronostiques et prédictives de la réponse aux traitements est capitale dans l’amélioration de la prise en charge de ce fléau mondial. Nous avons choisi une technique récente qui est la spectroscopie RMN HRMAS et des méthodes d’analyse statistique robustes (PCA et PLS-DA), afin d’établir les profils métaboliques des cancers épithéliaux de l’ovaire et des cancers mammaires. Nous avons établi, après un rappel théorique de la spectroscopie RMN HRMAS, un état d’art des applications médicales de cette technique, notamment gynécologiques chez la femme et urogénitales chez l’homme. Nous avons décrit les différentes étapes de la démarche établie pour l’analyse spectrale : préparation de l’échantillon tissulaire, acquisition RMN et analyse statistique. Nous avons montré que cette technique, permettant une analyse rapide (20 min) et non destructive d’échantillons tissulaires intacts, est applicable à la prise en charge thérapeutique des patientes atteintes de carcinomes ovariens et mammaires. Elle a permis, dans le cas des cancers de l’ovaire, de caractériser métaboliquement les trois types histologiques (séreux, endométrioïdes et mucineux) et le tissu ovarien sain, de générer des modèles statistiques permettant de classer les tumeurs borderline et de prédire la survie des patientes et la réponse à la chimiothérapie. Dans le cas des cancers du sein, elle a permis de discriminer métaboliquement les carcinomes mammaires, les fibroadénomes et le tissu sain et d’étudier métaboliquement les différents indicateurs histologiques des ces carcinomes. Nous projetons de confirmer ces résultats préliminaires très encourageants sur une plus grande cohorte. / Cancer is the leading cause of morbidity and mortality worldwide. The search for diagnostic, prognostic and predictive biomarkers of response to treatment is crucial in improving the management of this global scourge. We chose a new technique that is HRMAS NMR spectroscopy and robust statistical analysis methods (PCA and PLS-DA), to establish the metabolic profiles ofepithelial ovarian and breast cancers. We have determined, after a theoretical reminder of HRMAS NMR spectroscopy, a state of the art including medical applications of this technique, mainly gynecological in woman and uro-genital in man. We describe the different steps of the process established for spectral analysis : preparation of tissue sample, NMR acquisition and statistical analysis. We showed that this technique, allowing a rapid analysis (20 min) and non-destructive of intact tissue samples, is applicable to the therapeutic management of patients with breast and ovarian carcinomas. It has, in the case of ovarian cancer, characterize metabolically the three histological types (serous, endometrioid and mucinous) and healthy ovarian tissue, generate statistical models to classify borderline tumors and predict survival patients and response to chemotherapy. In the case of breast cancer, it could discriminate metabolically breast carcinomas, fibroadenomas and healthy tissue and study metabolically different histological indicators of these carcinomas. We plan to confirm these very encouraging preliminary results in a larger cohort.

From day to day, the role of HRMAS (High-Resolution Magic Angle Sinning) Nuclear Magnetic Resonance Spectroscopy (NMRS) in medical diagnosis is increasing. This technique enables setting up metabolite profiles of ex vivo pathological and healthy tissue. Automatic spectrum quantitation enables monitoring of diseases. However for several metabolites, the values of chemical shifts of proton groups may slightly differ according to the micro-environment in the tissue or cells, in particular to its pH. This hampers accurate estimation of the metabolite concentrations mainly when using quantitation algorithms based on a metabolite basis-set. The present word is devoted to the optimization of NMR metabolite basis set signals, particularly to the algorithms of chemical shift mismatch correction. Two sighal processing ("warping") methods were developed for simple and fast spectrum optimization : signal stretching/shrinking (resampling) and spectrum splitting. Then, another optimization method, QM-QUEST, coupling Quantrum Mechanical simulation and quantitation algorithms was implemented. The latter provides more robust fitting while limiting user involvement and respects the correct fingerprints of metabolites. Its efficiency is demonstrated by accurately quantitating signals from tissue samples of human brains with oligodendroglioma, obtained at 11.7 Tesla and spectra of cells acquired at 9.4T by HRMAS-NMR. As the necessity of fast NMR signal simulation based on quantum Mechanics is raised in the thesis, a part of the word is dedicated to an approximate method speeding-up the calculations. The algorithm based on spin-system fragmentation could become an important part of the QM-QUEST optimization method and will be implemented as an option of simulation in NMR-SCOPE, module of the jMRUI software package.

A few abnormal cells found in a small piece of prostate tissue are most consequential for a man’s future. The prevalence of prostate cancer (PCa) is increasing globally. The main instigating factor for this cancer is not yet known, but it appears to be the consequence of many variables such as an increasingly older population, more frequent PSA-testing, and factors involving lifestyle. Prostate cancer screening, as an equivalent for breast cancer screening, has been suggested but unfortunately there are no accurate diagnostic tools available for this type of screening. The reason for this is simply that the prostate is one of the most difficult organs to diagnose and, consequently, PCa screening would generate far too many false-positive and false-negative results.  The prostate is not easily accessible as it is deeply-seated in the male pelvic area, wrapped around the urethra and surrounded by sensitive vital organs.  Furthermore, PCa is frequently multi-focal, and the cancer cells have a tendency of assimilating among normal cells and, thus, do not always form solid lumps.  Therefore, prostate tumors are often not felt by digital rectal examination (DRE) or identified by imaging.  The PSA-test is not reliable as it is more prostate-specific than cancer-specific.  Due to increasing prostate awareness, more early-stage and locally confined PCa are being detected. This is saving lives, although there is a high risk of over treatment and unnecessary side-effects.  The increased detection of PCa requires sophisticated diagnostic methods and highly skilled clinicians who can discern between indolent and aggressive cancers.  The current “gold-standard” for PCa diagnosis is biopsy grading by pathologists using the Gleason score system, which is a difficult task.  Therefore, innovative methods to improve the precision of prostate diagnosis, by increased biopsy sensitivity and tumor localization, are of essence. In light of these difficulties, the metabolomic approach using 1D and 2D high-resolution magic angle spinning (HRMAS) NMR spectroscopy combined with histopathology on intact prostatectomy specimens was evaluated in this research project.  The non-destructive nature of HRMAS NMR enables spectroscopic analysis of intact tissue samples with consecutive histological examinations under light microscope. Metabolomics aids in the unraveling and the discovery of organ-specific endogenous metabolites that have the potential to be reliable indicators of organ function and viability, extrinsic and intrinsic perturbations, as well as valuable markers for treatment response. The results may, therefore, be applied clinically to characterize an organ by utilizing biomarkers that have the capacity to distinguish between disease and health. The aim was to characterize the human and the rat prostate in terms of its intermediary metabolism, which I show here to differ between species and anatomical regions.  Furthermore, the aim is to seek the verification of HRMAS NMR derived metabolites which are known to be a part of the prostate metabolome such as, citrate, choline, and the polyamines which were performed, but also the identification of metabolites not previously identified as part of the local prostate metabolism, such as Omega-6, which was detected in tumors.  The extended aim was to elucidate novel bio-markers with clinical potential. In this study, the common phyto-nutrient, inositol, which appears to possess protective properties, was identified as being a potentially important PCa bio-marker for the distinction between the more indolent Gleason score 6 and the more aggressive Gleason score 7 in non-malignant prostate tissues with tumors elsewhere in the organ. Further studies in this area of PCa research are therefore warranted.

La spectroscopie de Résonance Magnétique Nucléaire (RMN) Haute Résolution à l’angle magique (HRMAS) joue un rôle de plus en plus prépondérant pour le diagnostic médical. Cette technique permet d’établir les empreintes ex vivo des métabolites de tissus sains et pathologiques. Cependant, pour certains métabolites, les valeurs des déplacements chimiques des groupes de protons peuvent légèrement varier en fonction de l’environnement des tissus ou cellules, particulièrement de son acidité. Cet effet gêne l’estimation correcte des concentrations des métabolites lorsqu’on utilise des algorithmes fondés sur des bases de métabolites. Ce travail est dévolu aux méthodes d’optimisation des bases de métabolites, notamment aux algorithmes de correction des changements de déplacements chimiques. Deux méthodes de traitement du signal ont été développées pour l’optimisation simple et rapide des signaux / spectres : contraction/expansion du signal moyennant ré-échantillonnage et fractionnement du spectre. Une autre méthode, QM-QUEST, conjuguant la simulation par Mécanique Quantique et la quantification, a été mise en œuvre. Cette dernière permet l’ajustement plus robuste des spectres en limitant l’implication de l’utilisateur et préserve les empreintes correctes des métabolites. Son efficacité est démontrée pour la quantification de spectres RMN de biopsies cérébrales humaines d’oligodendroglioma, obtenues à 11.7 Tesla et de spectres de cellules acquis à 9.4 T par la technique RMN-HRMAS. Etant donné la nécessité de simulation rapide des signaux RMN basée sur la Mécanique Quantique, une partie du travail est vouée à une méthode approchée accélérant la simulation. L’algorithme fondé sur la fragmentation du système de spins pourrait devenir une partie importante de la méthode d’optimisation QM-QUEST et sema mis en œuvre en tant qu’option de simulation de la méthode NMR-SCOPE, module du logiciel jMRUI. / From day to day, the role of HRMAS (High-Resolution Magic Angle Sinning) Nuclear Magnetic Resonance Spectroscopy (NMRS) in medical diagnosis is increasing. This technique enables setting up metabolite profiles of ex vivo pathological and healthy tissue. Automatic spectrum quantitation enables monitoring of diseases. However for several metabolites, the values of chemical shifts of proton groups may slightly differ according to the micro-environment in the tissue or cells, in particular to its pH. This hampers accurate estimation of the metabolite concentrations mainly when using quantitation algorithms based on a metabolite basis-set. The present word is devoted to the optimization of NMR metabolite basis set signals, particularly to the algorithms of chemical shift mismatch correction. Two sighal processing (“warping”) methods were developed for simple and fast spectrum optimization : signal stretching/shrinking (resampling) and spectrum splitting. Then, another optimization method, QM-QUEST, coupling Quantrum Mechanical simulation and quantitation algorithms was implemented. The latter provides more robust fitting while limiting user involvement and respects the correct fingerprints of metabolites. Its efficiency is demonstrated by accurately quantitating signals from tissue samples of human brains with oligodendroglioma, obtained at 11.7 Tesla and spectra of cells acquired at 9.4T by HRMAS-NMR. As the necessity of fast NMR signal simulation based on quantum Mechanics is raised in the thesis, a part of the word is dedicated to an approximate method speeding-up the calculations. The algorithm based on spin-system fragmentation could become an important part of the QM-QUEST optimization method and will be implemented as an option of simulation in NMR-SCOPE, module of the jMRUI software package.

No description available.

Ces travaux de thèse portent sur l'analyse des mélanges réels et synthétiques complexes composés de petites molécules à l'aide de la RMN HRMAS. Dans une première partie, une approche RMN HRMAS basée sur l'analyse métabolomique en combinaison avec des techniques de reconnaissance des formes (PCA et O-PLS-DA) a été appliquée pour le diagnostic des lésions thyroïdiennes indéterminées et étudier également les effets biologiques négatifs des nanoparticules d'aluminium sur pseudomonas brassicacearum. Dans une seconde partie, nous avons étudié la RMN chromatographique en utilisant la silice comme matrice de support qui pourrait fournir une alternative rapide et complète de la LC pour la caractérisation de mélanges complexes. En outre, l'exigence de la suppression du signal dans l'extrait de plantes naturelles et d'hydrocarbures aromatiques conduit à l'élaboration d'une méthode rapide et précise en utilisant des polymères à empreintes moléculaires avec une excellente sélectivité. La sélectivité des polymères à empreintes moléculaires à travers la capture d'une cible moléculaire spécifique est exploitée ici pour éliminer efficacement les signaux RMN. / This thesis work deals with the analysis of natural and synthetic complex mixtures composed of small molecules using HRMAS NMR. In a first part, an integrated HRMAS-NMR based metabolomic analysis in combination with pattern recognition techniques (PCA and O-PLS-DA) has been applied for the diagnosis of indeterminate thyroid lesions and also studied the potential adverse biological effects of aluminium nanoparticles on pseudomonas brassicacearum. In a second part we investigated that chromatographic NMR using silica as the matrix support could provide a quick alternative and complement to LC for the characterization of complex mixtures. In addition, requirement for signal suppression in natural plant extract and aromatic hydrocarbons led to the development of a rapid and accurate method using molecularly imprinted polymers with excellent selectivity. The selectivity of Molecularly Imprinted polymers towards capturing a specific molecular target is exploited here to efficiently remove NMR signals.