Les études de perfusion du myocarde en tomographie par émission de positons (TEP) sont utilisées pour diagnostiquer l’étendue de maladies coronariennes. La TEP peut aussi mesurer le débit sanguin myocardique en valeur absolue, au repos et en période d’effort. Une des limitations réside dans le fait que l’activité résiduelle de l’examen de repos ne doit pas contaminer l’examen réalisé en phase d’effort. Il est donc nécessaire de séparer les deux acquisitions de 3 à 5 périodes radioactives. Il est maintenant possible d’utiliser de nouveaux traceurs marqués au fluor 18 pour quantifier le débit sanguin myocardique avec précision. Leur période offre la possibilité d’une distribution régionale mais complique la logistique des examens, en rallongeant la durée d’attente entre les deux acquisitions. Pour s’affranchir de cette contrainte, une nouvelle méthode a récemment été proposée (Alpert et al. 2012). Cette méthode permet de mesurer le débit sanguin myocardique au repos et en phase d’effort lors d’une seule session d’imagerie. Le but de cette thèse était de démontrer la faisabilité expérimentale de cette technique et de l’améliorer pour une éventuelle application clinique. Nous avons validé la méthode dans un modèle animal (cochon) pour le traceur 18F-Flurpiridaz puis, nous avons développé un algorithme pour générer rapidement des images paramétriques et des images standard de perfusion au repos et pendant effort. Nous avons ensuite validé nos méthodes pour le traceur 13NH3 qui est utilisé en clinique dans certains pays. Enfin, nous avons proposé une méthode pour la création d’un atlas multimodal 4D dans le but de faciliter les comparaisons entre acquisitions et sujets. / Positron emission tomography (PET) cardiac perfusion studies are used in the clinic to diagnose the extent of coronary artery disease. PET can also measure absolute myocardial blood flow (MBF) at rest and during stress. One limitation is that radioactivity from the rest scan must not affect the stress scan, making it necessary to wait 3 to 5 half-lives between studies for sufficient radioactive decay to occur. With the advent of new myocardial flow tracers, it is now possible to use 18F- tracers for myocardial perfusion imaging and accurate MBF quantification. Their physical half-life does not require the need of on-site cyclotron but constitutes a logistical complication for rest/stress measurements because of the between scan waiting period needed. To address this issue an alternative protocol was previously proposed (Alpert et al. 2012). In the new method rest and stress MBF is measured during a single-scan session. The goal of this thesis was to demonstrate the feasibility of this technique in experimental measurements and to improve its practicality for future clinical application. First, we demonstrated the use of the method in a porcine model for standard bull’s eye, segmental analysis using the novel flow tracer 18F-Flurpiridaz. We then extended the method and developed a computationally efficient algorithm to provide rest/stress MBF parametric maps and to generate rest/stress standard perfusion images in a timely manner. Third, we demonstrated the applicability of our methods to the clinically used radiotracer 13NH3. Lastly, we developed a framework for the construction of a 4D multimodal probabilistic atlas to facilitate inter- and intra-subjects comparisons.

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第19403号 / 医博第4054号 / 新制||医||1012(附属図書館) / 32428 / 京都大学大学院医学研究科医学専攻 / (主査)教授 前川 平, 教授 平岡 眞寛, 教授 羽賀 博典 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

No description available.

Phosphodiesterase 2A (PDE2A) is highly expressed in distinct areas of the brain, which are known to be related to neuropsychiatric diseases. The development of suitable PDE2A tracers for Positron Emission Tomography (PET) would permit the in vivo imaging of the PDE2A and evaluation of disease-mediated alterations of its expression. A series of novel fluorinated PDE2A inhibitors on the basis of a Benzoimidazotriazine (BIT) scaffold was prepared leading to a prospective inhibitor for further development of a PDE2A PET imaging agent. BIT derivatives (BIT1–9) were obtained by a seven-step synthesis route, and their inhibitory potency towards PDE2A and selectivity over other PDEs were evaluated. BIT1 demonstrated much higher inhibition than other BIT derivatives (82.9% inhibition of PDE2A at 10 nM). BIT1 displayed an IC50 for PDE2A of 3.33 nM with 16-fold selectivity over PDE10A. This finding revealed that a derivative bearing both a 2-fluoro-pyridin-4-yl and 2-chloro-5-methoxy-phenyl unit at the 8- and 1-position, respectively, appeared to be the most potent inhibitor. In vitro studies of BIT1 using mouse liver microsomes (MLM) disclosed BIT1 as a suitable ligand for 18F-labeling. Nevertheless, future in vivo metabolism studies are required.

Background: Boron Neutron Capture Therapy (BNCT) is a binary approach to cancer therapy that requires accumulation of boron atoms preferentially in tumour cells. This can be achieved by using nanoparticles as boron carriers and taking advantage of the enhanced permeability and retention (EPR) effect. Here, we present the preparation and characterization of size and shape-tuned gold NPs (AuNPs) stabilised with polyethylene glycol (PEG) and functionalized with the boron-rich anion cobalt bis(dicarbollide), commonly known as COSAN. The resulting NPs were radiolabelled with 124I both at the core and the shell, and were evaluated in vivo in a mouse model of human fibrosarcoma (HT1080 cells) using positron emission tomography (PET). Methods: The thiolated COSAN derivatives for subsequent attachment to the gold surface were synthesized by reaction of COSAN with tetrahydropyran (THP) followed by ring opening using potassium thioacetate (KSAc). Iodination on one of the boron atoms of the cluster was also carried out to enable subsequent radiolabelling of the boron cage. AuNPs grafted with mPEG-SH (5 Kda) and thiolated COSAN were prepared by ligand displacement. Radiolabelling was carried out both at the shell (isotopic exchange) and at the core (anionic absorption) of the NPs using 124I to enable PET imaging. Results: Stable gold nanoparticles simultaneously functionalised with PEG and COSAN (PEG-AuNPs@[4]) with hydrodynamic diameter of 37.8 0.5 nm, core diameter of 19.2 1.4 nm and -potential of 18.0 0.7 mV were obtained. The presence of the COSAN on the surface of the NPs was confirmed by Raman Spectroscopy and UV-Vis spectrophotometry. PEG-AuNPs@[4] could be efficiently labelled with 124I both at the core and the shell. Biodistribution studies in a xenograft mouse model of human fibrosarcoma showed major accumulation in liver, lungs and spleen, and poor accumulation in the tumour. The dual labelling approach confirmed the in vivo stability of the PEG-AuNPs@[4]. Conclusions: PEG stabilized, COSAN-functionalised AuNPs could be synthesized, radiolabelled and evaluated in vivo using PET. The low tumour accumulation in the animal model assayed points to the need of tuning the size and geometry of the gold core for future studies.

No description available.

Background: Alterations of hypothalamic–pituitary–adrenal (HPA) axis activity and serotonergic signaling are implicated in the pathogenesis of human obesity and may contribute to its metabolic and mental complications. The association of these systems has not been investigated in human obesity. Objective: To investigate the relation of HPA responsiveness and serotonin transporter (5-HTT) availability in otherwise healthy individuals with obesity class II or III (OB) compared to non-obesity controls (NO). Study participants: Twenty-eight OB (21 females; age 36.6 10.6 years; body mass index (BMI) 41.2 5.1 kg/m2) were compared to 12 healthy NO (8 females; age 35.8 7.4 years; BMI 22.4 2.3 kg/m2), matched for age and sex. Methods: HPA axis responsiveness was investigated using the combined dexamethasone/corticotropin-releasing hormone (dex/CRH) test, and curve indicators were derived for cortisol and adrenocorticotropic hormone (ACTH). The 5-HTT selective tracer [11C]DASB was applied, and parametric images of the binding potentials (BPND) were calculated using the multilinear reference tissue model and evaluated by atlas-based volume of interest (VOI) analysis. The self-questionnaires of behavioral inhibition system/behavioral activation system (BIS/BAS) with subscales drive, fun-seeking and reward were assessed. Results: OB showed significant positive correlations of ACTH curve parameters with overall 5-HTT BPND (ACTHAUC: r = 0.39, p = 0.04) and 5-HTT BPND of the caudate nucleus (ACTHAUC: r = 0.54, p = 0.003). In NO, cortisol indicators correlated significantly with BPND in the hippocampus (cortisolAUC: r = 0.59, p = 0.04). In OB, BAS reward was inversely associated with the ACTHAUC (r = 0.49, p = 0.009). Conclusion: The present study supports a serotonergic-neuroendocrine association, which regionally differs between OB and NO. In OB, areas processing emotion and reward seem to be in-volved. The finding of a serotonergic HPA correlation may have implications for other diseases with dysregulated stress axis responsiveness, and for potential pharmacologic interven-tions.

Eine der drei Säulen der Krebsbehandlung ist die Strahlentherapie. Einer der neuesten Ansätze hierbei ist die Bestrahlung mit Ionen, zurzeit insbesondere Protonen und Kohlenstoffionen. Diese Hochpräzisionstherapie erfordert ein hohes Maß an Kontrolle, da die applizierte Dosisverteilung sehr empfindlich von Dichteveränderungen im durchstrahlten Gewebe abhängt. Das bisher einzige klinisch eingesetzte Verfahren zur in vivo Überwachung der Dosisapplikation bei Ionenbestrahlungen ist die Positronen-Emissions-Tomographie (PET). Sie ermöglicht eine Verifikation der Teilchenreichweite sowie der Lage des Bestrahlungsfeldes. Die mit der PET-Methode gemessene Aktivitätsverteilung lässt sich jedoch nicht direkt mit der geplanten Dosisverteilung vergleichen. Daher ist eine Vorherberechnung der erwarteten Aktivitätsverteilung auf der Grundlage des Bestrahlungsplanes notwendig, welche dann mit der Messung verglichen wird und eine qualitative Beurteilung der Bestrahlung ermöglicht. Die Vorherberechnung der erwarteten Aktivitätsverteilung erfordert bislang die Kenntnis einer Vielzahl von Wirkungsquerschnitten. Nur für wenige dieser Wirkungsquerschnitte liegen jedoch Messdaten im benötigten Energiebereich und mit ausreichender Genauigkeit vor. Daher verwenden viele Monte-Carlo-Simulationen intrinsische Kernmodelle oder semi-empirische Modellierungen, die häufig eine unzureichende Genauigkeit aufweisen. In Fachkreisen ist bisher noch nicht geklärt, welches die optimale Ionensorte für die Tumortherapie ist. Insbesondere Lithiumionen weisen aufgrund ihrer physikalischen und radiobiologischen Eigenschaften ein großes Potenzial auf. Auch für Bestrahlungen mit diesen Ionen ist ein PET-Monitoring der Therapie erstrebenswert. In der vorliegenden Arbeit wird zunächst die Anwendbarkeit der Reichweite-Verifikation mittels PET bei Bestrahlung mit Lithiumionen gezeigt. Des Weiteren wird ein Konzept zur Modellierung der Positronenemitter-Verteilung ohne Kenntnis der Wirkungsquerschnitte entwickelt. Diese Vorhersage beruht auf in Referenzmaterialien (Wasser, Graphit und Polyethylen) gemessenen tiefenabhängigen Positronenemitter-Yields, mit welchen durch geeignete Linearkombination die Verteilung der Positronenemitter in beliebigen Materialien bekannter Stöchiometrie vorausberechnet werden kann. Die Anwendbarkeit des Yield-Konzeptes wird gezeigt für Lithium- und Kohlenstoffbestrahlungen homogener Polymethylmethacrylat (PMMA) Targets sowie verschiedener inhomogener Targets. / One of the three main tumour treatment forms is radiation therapy. Here, the application of ion beams, in particular protons and carbon ions, is of growing importance. This high precision therapy requires a consequent monitoring of the dose delivery since the induced dose deposition is very sensitive to density changes in the irradiated tissue. Up to now, positron emission tomography (PET) is the only in vivo method in clinical use for monitoring the dose deposition in ion beam therapy. It allows for the verification of the particle range as well as the position of the irradiation field. The distribution of activity measured by means of PET cannot be compared directly to the planned dose distribution. Thus, a calculation of the expected activity distribution is required which then can be compared to the measurement. Simulation of the expected activity distribution requires the exact knowledge of various cross sections. Only a few of them have been measured in the required energy range so far. Therefore, in Monte Carlo simulations often intrinsic nuclear models or semi-empirical parametrization are used which often exhibit insufficient accuray. Among experts the question on the optimum ion species for tumour therapy is still open. Especially lithium ions exhibit a great potential due to their favourable physical and radiobiological properties. Also for these ions a PET monitoring is highly desirable. The presented work shows the feasibility of range verification by means of PET for lithium irradiation. Furthermore, a concept for modeling positron emitter distributions without the knowledge of cross sections is developed. This prediction is based on depth-dependent positron emitter yields measured in reference materials (water, graphite and polyethylene). With these data the positron emitter distribution in any material of known stoichiometry can be calculated by means of an appropriate linear combination. The feasibility of the yield concept is shown for lithium and carbon irradiation, respectively, of homogeneous polymethyl methacrylate (PMMA) as well as various inhomogeneous targets.

Eine der drei Säulen der Krebsbehandlung ist die Strahlentherapie. Einer der neuesten Ansätze hierbei ist die Bestrahlung mit Ionen, zurzeit insbesondere Protonen und Kohlenstoffionen. Diese Hochpräzisionstherapie erfordert ein hohes Maß an Kontrolle, da die applizierte Dosisverteilung sehr empfindlich von Dichteveränderungen im durchstrahlten Gewebe abhängt. Das bisher einzige klinisch eingesetzte Verfahren zur in vivo Überwachung der Dosisapplikation bei Ionenbestrahlungen ist die Positronen-Emissions-Tomographie (PET). Sie ermöglicht eine Verifikation der Teilchenreichweite sowie der Lage des Bestrahlungsfeldes. Die mit der PET-Methode gemessene Aktivitätsverteilung lässt sich jedoch nicht direkt mit der geplanten Dosisverteilung vergleichen. Daher ist eine Vorherberechnung der erwarteten Aktivitätsverteilung auf der Grundlage des Bestrahlungsplanes notwendig, welche dann mit der Messung verglichen wird und eine qualitative Beurteilung der Bestrahlung ermöglicht. Die Vorherberechnung der erwarteten Aktivitätsverteilung erfordert bislang die Kenntnis einer Vielzahl von Wirkungsquerschnitten. Nur für wenige dieser Wirkungsquerschnitte liegen jedoch Messdaten im benötigten Energiebereich und mit ausreichender Genauigkeit vor. Daher verwenden viele Monte-Carlo-Simulationen intrinsische Kernmodelle oder semi-empirische Modellierungen, die häufig eine unzureichende Genauigkeit aufweisen. In Fachkreisen ist bisher noch nicht geklärt, welches die optimale Ionensorte für die Tumortherapie ist. Insbesondere Lithiumionen weisen aufgrund ihrer physikalischen und radiobiologischen Eigenschaften ein großes Potenzial auf. Auch für Bestrahlungen mit diesen Ionen ist ein PET-Monitoring der Therapie erstrebenswert. In der vorliegenden Arbeit wird zunächst die Anwendbarkeit der Reichweite-Verifikation mittels PET bei Bestrahlung mit Lithiumionen gezeigt. Des Weiteren wird ein Konzept zur Modellierung der Positronenemitter-Verteilung ohne Kenntnis der Wirkungsquerschnitte entwickelt. Diese Vorhersage beruht auf in Referenzmaterialien (Wasser, Graphit und Polyethylen) gemessenen tiefenabhängigen Positronenemitter-Yields, mit welchen durch geeignete Linearkombination die Verteilung der Positronenemitter in beliebigen Materialien bekannter Stöchiometrie vorausberechnet werden kann. Die Anwendbarkeit des Yield-Konzeptes wird gezeigt für Lithium- und Kohlenstoffbestrahlungen homogener Polymethylmethacrylat (PMMA) Targets sowie verschiedener inhomogener Targets. / One of the three main tumour treatment forms is radiation therapy. Here, the application of ion beams, in particular protons and carbon ions, is of growing importance. This high precision therapy requires a consequent monitoring of the dose delivery since the induced dose deposition is very sensitive to density changes in the irradiated tissue. Up to now, positron emission tomography (PET) is the only in vivo method in clinical use for monitoring the dose deposition in ion beam therapy. It allows for the verification of the particle range as well as the position of the irradiation field. The distribution of activity measured by means of PET cannot be compared directly to the planned dose distribution. Thus, a calculation of the expected activity distribution is required which then can be compared to the measurement. Simulation of the expected activity distribution requires the exact knowledge of various cross sections. Only a few of them have been measured in the required energy range so far. Therefore, in Monte Carlo simulations often intrinsic nuclear models or semi-empirical parametrization are used which often exhibit insufficient accuray. Among experts the question on the optimum ion species for tumour therapy is still open. Especially lithium ions exhibit a great potential due to their favourable physical and radiobiological properties. Also for these ions a PET monitoring is highly desirable. The presented work shows the feasibility of range verification by means of PET for lithium irradiation. Furthermore, a concept for modeling positron emitter distributions without the knowledge of cross sections is developed. This prediction is based on depth-dependent positron emitter yields measured in reference materials (water, graphite and polyethylene). With these data the positron emitter distribution in any material of known stoichiometry can be calculated by means of an appropriate linear combination. The feasibility of the yield concept is shown for lithium and carbon irradiation, respectively, of homogeneous polymethyl methacrylate (PMMA) as well as various inhomogeneous targets.

Résumé : Les personnes atteintes de la maladie d’Alzheimer présentent une diminution de la capture cérébrale du glucose, qui semble être impliquée dans le développement des problèmes cognitifs associés à la maladie. Toutefois, il n’y a encore aucun consensus quant à savoir si la capture cérébrale du glucose est diminuée chez les personnes âgées cognitivement saines. En condition de déficit de glucose, les cétones sont le substrat énergétique alternatif pour le cerveau. La diète cétogène, induisant une cétose légère, améliore les fonctions cognitives chez les modèles animaux et chez l’homme. Notre premier objectif était d’évaluer l’effet du vieillissement sain et de la diète cétogène sur la capture cérébrale du glucose et des cétones chez le rat. Pour cela, la capture cérébrale de radiotraceurs analogues au glucose et aux cétones a été mesurée par tomographie par émission de positons. Nos résultats montrent que la capture des deux principaux substrats énergétiques du cerveau est globalement similaire chez des rats sains jeunes et âgés, mais est plus élevée suite à la diète cétogène. L’induction d’une cétose légère pourrait corriger la diminution de capture cérébrale du glucose subvenant au cours de la maladie d’Alzheimer. Le second objectif était de déterminer l’effet d’une diète cétogène sur le métabolisme cérébral du glucose et des cétones chez le rat. Pour cela, les différents intermédiaires des voies métaboliques du glucose et des cétones ont été mesurés par spectroscopie par résonance magnétique nucléaire. Les résultats démontrent que le métabolisme du glucose et des cétones dans les cellules du cerveau est plus élevé suite à la diète cétogène. Le contenu en acide [gamma]-aminobutyrique, le principal neurotransmetteur inhibiteur, est aussi plus élevé suite à la diète cétogène, ce qui pourrait contribuer à l’effet antiépileptique de la diète cétogène. Le troisième objectif était d’évaluer l’impact d’une restriction calorique à long terme, pouvant induire une cétose légère, sur le métabolisme cérébral chez des rats âgés sains. Nos résultats montrent que, couplée à une diète à haute teneur en sucrose et faible en acides gras oméga-3, la restriction calorique à long terme chez les rats âgés ne modifie pas le profil des métabolites et des acides gras du cerveau. La déficience en acides gras oméga-3 et la surcharge de sucrose pourraient empêcher une grande partie des effets bénéfiques de la restriction calorique au cerveau.//Abstract : Alzheimer’s disease is associated with a reduction of brain glucose uptake, which may be involved in the development of the cognitive problems associated with the disease. It is however unclear whether brain glucose uptake is decreased in the cognitively healthy elderly. Under conditions of glucose deficit, ketones are the alternative brain energy substrate. A mild ketosis, induced by the ketogenic diet, improves cognitive functions in animal models and humans. Our first objective was to evaluate the effect of healthy aging and of a ketogenic diet on brain glucose and ketone uptake in the rat. Brain uptake of radiotracers analogous to glucose and ketones was measured by positron emission tomography. Our results show that the uptake of the brains two main energy substrates is generally similar in healthy young and aged rats, but is higher under the ketogenic diet. The induction of a mild ketosis may compensate the reduction of brain glucose uptake occurring in Alzheimer’s disease. The second objective was to assess the effect of a ketogenic diet on brain glucose and ketone metabolism in the rat. Metabolic pathway intermediates of glucose and ketones were measured by nuclear magnetic resonance spectroscopy. Results show that glucose and ketone metabolism in brain cells is higher under the ketogenic diet. Content of [gamma]-aminobutyric acid, the main inhibitory neurotransmitter, is also higher under the ketogenic diet, which could contribute to the antiepileptic effect of the ketogenic diet. The third objective was to evaluate the effect of a long-term calorie restriction, which may induce a mild ketosis, on brain metabolism in healthy aged rats. Our results show that, in conjunction with a diet enriched in sucrose and low in omega-3 fatty acids, long-term calorie restriction in aged rats does not change brain metabolite and fatty acid profiles. Omega-3 fatty acid deficiency and an overload of sucrose may prevent the beneficial effects associated with calorie restriction in the brain.