Thesis (M.A.)--Boston University / The standardized uptake value (SUV) is increasingly being used for diagnosis, staging, and monitoring disease in clinical oncology. Comparing tumor SUV to background SUV is an attractive way to minimize variability and ensure the quality of scans across different institutions. The liver has been identified as a potential source for background normalization, however no studies have compared the liver to other background sites for a variety of cancers. The purpose of this study was to evaluate the use of liver uptake for the standardization of FDG PET/CT imaging. Scans from 145 patients were prospectively reviewed under the supervision of a radiologist with board certification in nuclear medicine (R.M.S. , 3 years of experience). Liver SUV values were correlated to mediastinum SUV values in lung and breast cancer patients, and internal jugular vein (IJV) SUV values in head and neck cancer patients. The independent t-test was used to determine if there was a statistically significant affect of the amount of incubation time or use of intravenous contrast on the SUV. For the lung and breast cancer patients, a strong correlation was observed between the mediastinum SUVmean and liver SUVmean (r = 0.89), whereas for the head and neck cancer patients, a weaker correlation was observed between the IJV SUVmean and the liver SUVmean (r = 0.69). Neither the amount of incubation time nor the use of IV contrast demonstrated a significant affect on the SUV. We conclude that liver SUVmean may be used to standardize FOG PET/CT studies in cancers of the lung, breast and head and neck. However, additional studies in other cancers as well as the affects of age, gender, benign disease and use of chemotherapy are still desired before widespread adoption of this standard.

Poly(ADP-ribose)polymerase-1 and -2 (PARP1/2) are nuclear proteins involved in DNA repair. Tumors with defects in homologous recombination, including BRCA1- and BRCA2-deficient cancers, have been shown to be sensitive to PARP inhibition. The Weissleder group has synthesized fluorescent and radioactive derivatives of the PARP1/2 inhibitor AZD2281. We hypothesized that fluorescent and radioactive AZD2281-based imaging agents would quantify PARP1/2 expression in vitro and in vivo. To test this hypothesis, a panel of pancreatic ductal adenocarcinoma and ovarian carcinoma cell lines were characterized by immunocytochemistry for PARP1/2 expression. AZD2281-derived fluorescence signal correlated with anti-PARP antibody fluorescence signal strength in vitro. Four cell lines representing a range of PARP1/2 expression levels were then xenografted into Nu/Nu mice. Mice bearing four tumor types each were imaged with AZD2281-derived imaging agents, sacrificed, and their tumors excised for stand-alone imaging and Western blot. AZD2281-derived signal correlated with tumor PARP1/2 expression determined by Western blot, indicating that PARP1/2 expression level is a determinant of fluorescent signal strength and SUVs of AZD2281-derived agents in vivo. These data indicate that AZD2281-derived agents are useful tools for quantifying intracellular PARP1/2 both in vitro and in vivo, which could one day enable prospective identification of tumors likely to respond to PARP inhibitors.

Introduction. Fused Positron Emission Tomography and Computed Tomography (PET/CT) is an emerging investigative tool in cardiovascular disease that provides an imaging-based quantification of pathophysiological processes of interest. The purpose of this thesis was to study the application of PET to identify fundamental pathophysiological processes driving 3 forms of cardiovascular disease: aortic stenosis, myocardial infarction, and atherosclerosis. Methods. Aortic Stenosis. Patients with a spectrum of calcific aortic valve disease (n=121) who underwent PET-CT imaging for the identification of valvular calcification (18Ffluoride) and inflammation (18F-fluorodeoxyglucose, 18F-FDG) underwent serial imaging and clinical follow-up over 2 years. Baseline imaging findings were compared with echocardiographic and CT markers of disease progression and clinical outcome. Myocardial Infarction. Patients underwent PET-CT imaging with 18F-fluciclatide (a novel αvβ3-selective radiotracer highlighting active angiogenesis, inflammation and fibrosis) after ST-segment elevation MI (n=21), alongside stable patients with chronic total occlusion (CTO) of a major coronary vessel (n=7), and healthy volunteers (n=9). Myocardial radiotracer uptake was compared with clinical and cardiac magnetic resonance imaging (CMR) markers of infarction and remodeling. Atherosclerosis. Patients with a spectrum of atherosclerotic disease categorized as stable or unstable (recent MI) underwent PET/CT imaging with 18F-fluciclatide (n=46). Thoracic aortic 18F-fluciclatide uptake was compared with aortic atherosclerotic burden quantified by CT plaque thickness, plaque volume and calcium scoring. Histological validation. Tissue from the aortic valve, myocardium and carotid arteries of study subjects was acquired and examined ex vivo using histology and autoradiography. Results. Aortic Stenosis. Baseline valvular 18F-fluoride uptake correlated strongly with the rate of progression in AVC (r=0.80, p < 0.001) and with haemodynamic progression (mean aortic valve gradient r=0.32, p=0.001). It emerged as independently associated with clinical outcome after age and sex-adjustment (HR 1.55 [1.33-1.81], p < 0.001). 18F-FDG demonstrated moderate correlations with disease progression as assessed by CT (r=0.43, p=0.001) and echocardiography (18F-FDG r=0.30, p=0.001), and was associated with clinical outcomes independent of age and sex (HR 1.35 [1.16-1.58], p < 0.001). Valvular 18F-fluoride uptake correlated with immunohistochemical markers of calcification activity. There was no correlation between 18F-FDG uptake and inflammation. Myocardial Infarction. 18F-Fluciclatide binding was demonstrated in ex vivo peri-infarct myocardium and uptake was increased in vivo at sites of acute infarction compared to remote myocardium (tissue-to-background ratio (TBRmean) 1.34±0.22 vs 0.85±0.17 respectively, p < 0.001) and myocardium of healthy volunteers (TBRmean 1.34±0.22 vs 0.70±0.03; p < 0.001). There was no 18F-fluciclatide uptake at sites of established prior infarction in patients with CTO, with myocardial activity similar to healthy volunteers (TBRmean 0.71±0.06 vs. 0.70±0.03,p=0.83). 18F-Fluciclatide uptake occurred at sites of regional wall hypokinesia (wall motion index ≥1 vs 0; TBRmean 0.93±0.31 vs 0.80±0.26 respectively, p < 0.001), was increased in segments displaying functional recovery (TBRmean 0.95±0.33 vs 0.81±0.27, p=0.002) and associated with increase in probability of regional recovery. Atherosclerosis. 18F-Fluciclatide vascular binding ex vivo co-localised with regions of increased αvβ3 integrin expression, and markers of inflammation and angiogenesis. 18F-Fluciclatide uptake in vivo correlated with measures of aortic atherosclerotic burden: plaque thickness (r=0.57, p=0.001), total plaque volume (r=0.56, p=0.001) and the CT aortic calcium score (r=0.37, p=0.01). Patients with recent MI had greater aortic 18F-fluciclatide uptake than those with stable disease (TBRmax 1.33 vs 1.21, p=0.01). Conclusions. In a range of cardiovascular diseases, PET-CT can provide insights into key pathophysiological processes, guide patient risk stratification and prognosis, and identify important biomarkers of disease activity that can be used for the development of future therapeutic interventions.

La radiothérapie est une des méthodes principales utilisée dans le traitement thérapeutique des tumeurs malignes. Pour améliorer son efficacité, deux problèmes essentiels doivent être soigneusement traités : la prédication fiable des résultats thérapeutiques et la segmentation précise des volumes tumoraux. La tomographie d’émission de positrons au traceur Fluoro- 18-déoxy-glucose (FDG-TEP) peut fournir de manière non invasive des informations significatives sur les activités fonctionnelles des cellules tumorales. Les objectifs de cette thèse sont de proposer: 1) des systèmes fiables pour prédire les résultats du traitement contre le cancer en utilisant principalement des caractéristiques extraites des images FDG-TEP; 2) des algorithmes automatiques pour la segmentation de tumeurs de manière précise en TEP et TEP-TDM. La théorie des fonctions de croyance est choisie dans notre étude pour modéliser et raisonner des connaissances incertaines et imprécises pour des images TEP qui sont bruitées et floues. Dans le cadre des fonctions de croyance, nous proposons une méthode de sélection de caractéristiques de manière parcimonieuse et une méthode d’apprentissage de métriques permettant de rendre les classes bien séparées dans l’espace caractéristique afin d’améliorer la précision de classification du classificateur EK-NN. Basées sur ces deux études théoriques, un système robuste de prédiction est proposé, dans lequel le problème d’apprentissage pour des données de petite taille et déséquilibrées est traité de manière efficace. Pour segmenter automatiquement les tumeurs en TEP, une méthode 3-D non supervisée basée sur le regroupement évidentiel (evidential clustering) et l’information spatiale est proposée. Cette méthode de segmentation mono-modalité est ensuite étendue à la co-segmentation dans des images TEP-TDM, en considérant que ces deux modalités distinctes contiennent des informations complémentaires pour améliorer la précision. Toutes les méthodes proposées ont été testées sur des données cliniques, montrant leurs meilleures performances par rapport aux méthodes de l’état de l’art. / Radiation therapy is one of the most principal options used in the treatment of malignant tumors. To enhance its effectiveness, two critical issues should be carefully dealt with, i.e., reliably predicting therapy outcomes to adapt undergoing treatment planning for individual patients, and accurately segmenting tumor volumes to maximize radiation delivery in tumor tissues while minimize side effects in adjacent organs at risk. Positron emission tomography with radioactive tracer fluorine-18 fluorodeoxyglucose (FDG-PET) can noninvasively provide significant information of the functional activities of tumor cells. In this thesis, the goal of our study consists of two parts: 1) to propose reliable therapy outcome prediction system using primarily features extracted from FDG-PET images; 2) to propose automatic and accurate algorithms for tumor segmentation in PET and PET-CT images. The theory of belief functions is adopted in our study to model and reason with uncertain and imprecise knowledge quantified from noisy and blurring PET images. In the framework of belief functions, a sparse feature selection method and a low-rank metric learning method are proposed to improve the classification accuracy of the evidential K-nearest neighbor classifier learnt by high-dimensional data that contain unreliable features. Based on the above two theoretical studies, a robust prediction system is then proposed, in which the small-sized and imbalanced nature of clinical data is effectively tackled. To automatically delineate tumors in PET images, an unsupervised 3-D segmentation based on evidential clustering using the theory of belief functions and spatial information is proposed. This mono-modality segmentation method is then extended to co-segment tumor in PET-CT images, considering that these two distinct modalities contain complementary information to further improve the accuracy. All proposed methods have been performed on clinical data, giving better results comparing to the state of the art ones.