This thesis focuses on multilocus methods designed to detect single nucleotide polymorphisms (SNPs) that are associated with disease using case-control data. I study multilocus methods that allow for interaction in the regression model because epistasis is thought to be pervasive in the etiology of common human diseases. In contrast, the single-SNP models widely used in genome wide association studies (GWAS) are thought to oversimplify the underlying biology. I consider both pairwise interactions between individual SNPs and modular interactions between sets of biologically similar SNPs. Modular epistasis may be more representative of disease processes and its incorporation into regression analyses yields more parsimonious models. My methodological work focuses on strategies to increase power to detect susceptibility SNPs in the presence of genetic interaction. I emphasize the effect of gene-gene independence constraints and explore methods to relax them. I review several existing methods for interaction analyses and present their first empirical evaluation in a GWAS setting. I introduce the innovative retrospective Tukey score test (RTS) that investigates modular epistasis. Simulation studies suggest it offers a more powerful alternative to existing methods. I present diverse applications of these methods, using data from a multi-stage GWAS on prostate cancer (PRCA). My applied work is designed to generate hypotheses about the functionality of established susceptibility regions for PRCA by identifying SNPs that affect disease risk through interactions with them. Comparison of results across methods illustrates the impact of incorporating different forms of epistasis on inference about disease association. The top findings from these analyses are well supported by molecular studies. The results unite several susceptibility regions through overlapping biological pathways known to be disrupted in PRCA, motivating replication study.

Le cancer de la prostate (CaP) est le plus diagnostiqué chez les hommes au Canada et représente le troisième cancer le plus meurtrier au sein de cette population. Malgré l’efficacité des traitements de première ligne, de nombreux patients finiront par développer une résistance et, le cas échéant, verront leur CaP progresser vers une forme plus agressive. Plusieurs paramètres, essentiellement cliniques, permettent de prédire la progression du CaP mais leur sensibilité, encore limitée, implique la nécessité de nouveaux biomarqueurs afin de combler cette lacune. Dans cette optique nous nous intéressons au facteur de transcription NF-κB. Des études réalisées au laboratoire et ailleurs, associent RelA(p65) à un potentiel clinique dans le CaP, soulignant ainsi l’importance de la voie classique NF-κB. L’implication de la voie alternative NF-κB dans la progression du CaP a aussi été suggérée dans une de nos études illustrant la corrélation entre la distribution nucléaire de RelB et le score de Gleason. Alors que la voie classique est largement documentée et son implication dans la progression du CaP établie, la voie alternative, elle, reste à explorer. La présente thèse vise à clarifier l’implication de la voie alternative NF-κB dans le CaP et répond à deux objectifs fixés dans ce but. Le premier objectif fut d’évaluer l’impact de l'activation de la voie alternative NF-κB sur la biologie des cellules cancéreuses prostatiques. L’étude de la surexpression de RelB a souligné les effets de la voie alternative NF-κB sur la prolifération et l'autophagie. Étant ainsi impliquée tant dans la croissance tumorale que dans un processus de plus en plus associée à la progression tumorale, quoique potentiellement létal pour les cellules cancéreuses, son impact sur la tumorigénèse du CaP reste encore difficile à définir. Il n'existe, à ce jour, aucune étude permettant de comparer le potentiel clinique des voies classique et alternative NF-κB. Le second objectif de ce projet fut donc l'analyse conjointe de RelA(p65) et RelB au sein de mêmes tissus de patients atteints de CaP afin de déterminer l'importance clinique des deux signalisations NF-κB, l'une par rapport à l'autre. Le marquage immunofluorescent de RelA(p65) et RelB en a permis l'analyse quantitative et objective par un logiciel d'imagerie. Nos travaux ont confirmé le potentiel clinique associé à RelA(p65). La variable RelA(p65)/RelB s’est, elle, avérée moins informative que RelA(p65). Par contre, aucune corrélation entre RelB et les paramètres cliniques inclus dans l'étude n’est ressortie. En définitive, mon projet de thèse aura permis de préciser l'implication de la voie alternative NF-κB sur la biologie du CaP. Son impact sur la croissance des cellules cancéreuses prostatiques ainsi que sur l'autophagie, dénote l’ambivalence de la voie alternative NF-κB face à la tumorigénèse du CaP. L’étude exhaustive de la signalisation NF-κB souligne davantage l'importance de la voie classique dont l’intérêt clinique est principalement associé au statut de RelA(p65). Ainsi, bien que RelB n’affiche aucun potentiel en tant que biomarqueur exploitable en clinique, l’analyse de l’intervention de la voie alternative NF-κB sur la biologie des cellules cancéreuses prostatiques reste d’intérêt pour la compréhension de son rôle exact dans la progression du CaP. / Prostate cancer (PCa) is the most frequently diagnosed cancer and represents the third cause of cancer-death in Canadian men. Despite effective first-line therapies, many patients experience disease recurrence where PCa progresses toward a more aggressive form. Several parameters, largely clinical, have been used to predict the progression of PCa but their accuracy is still limited and implies the need for new biomarkers to fill this gap. Previous research from the laboratory has demonstrated that the transcription factor NF-B, and its nuclear localization, could be such a prognostic biomarker. Studies in our laboratory and elsewhere have correlated RelA(p65) with a clinical progression in PCa, underlining the importance of the classical NF-B pathway. The involvement of the alternative NF-B pathway in the progression of PCa was also suggested in one of our studies showing the correlation between the nuclear distribution of RelB and Gleason score. While the classical pathway is well documented and its involvement in the PCa progression established, the alternative NF-B pathway remains largely unexplored. This thesis describes two research objectives that aims to clarify the involvement of the alternative NF-B pathway in PCa. The first objective assessed the impact of the alternative NF-B pathway activation on the biology of PCa cells. RelB overexpression in 22RV1 PCa cells highlighted an effect of the alternative NF-B pathway on cell proliferation and autophagy. Its dual role in cell growth and a form of cell death requires further study to understand the balance of these in PCa tumorigenesis. To date no study has addressed the comparative prognostic potential of both the classical and alternative NF-B pathways simultaneously. Therefore the second objective of this research project was to analyze both RelA(p65) and RelB at a cellular level in the same tissue of patients with PCa to determine their unique and combined contribution to predicting biochemical recurrence in patients. This analysis was possible through immunofluorescent labeling of RelA (p65) and RelB, and was followed by a quantitative and objective analysis using an appropriate software. Our work confirmed the predictive value of RelA(p65) for biochemical recurrance. Combining RelA(p65) with RelB weakened the association, and RelB on its own was not found to predict biochemical recurrance. Ultimately, the research presented here has clarified the involvement of the alternative NF-B pathway on the biology of PCa. Its impact on the growth of PCa cells as well as autophagy reveals the dual role of the alternative NF-B pathway in PCa tumorigenesis. This exhaustive study of NF-B in PCa tissues further underscores the importance of the classical pathway whose clinical interest is mainly associated with RelA(p65) status. Thus, although RelB shows no potential as a clinically exploitable biomarker, further studies are needed to determine whether RelB contributes, either positively or negatively, and in a temporal fashion, to PCa progression.

The chemokine receptor CCR5 has been implicated in the pathogenesis of prostate cancer (PCa). A novel series of piperazine derivatives have been designed and synthesized as CCR5 antagonists and their activity as inhibitors of PCa cell lines proliferation was explored. A lead compound has been identified which induced 100% inhibition of PCa cell proliferation at micromolar concentrations. Anibamine, the only natural product CCR5 antagonist, was also examined for its anti-proliferative activity and was found to inhibit proliferation of PCa cells at micromolar concentrations as well. The expression of RANTES mRNA was observed in DU-145, M12 and P69 cells via RT-PCR, while the expression of CCR5 mRNA was observed only in M12 cells. A CHO-CCR5 stable cell line was prepared for the CCR5 ligand competition binding assays. Both anibamine and the newly identified lead compound will serve as leads in the development of novel CCR5 antagonists as anti-prostate cancer agents.

A fundamental challenge in the treatment planning process of multi-fractional external-beam radiation therapy (EBRT) is the tradeoff between tumor control and normal tissue sparing in the presence of geometric uncertainties (GUs). To accommodate GUs, the conventional way is to use an empirical planning treatment volume (PTV) margin on the treatment target. However, it is difficult to determine a near-optimal PTV margin to ensure specified target coverage with as much normal tissue protection as achievable. Coverage optimized planning (COP) avoids this problem by optimizing dose in possible virtual treatment courses with GU models directly incorporated. A near-optimal dosimetric margin generated by COP was reported to savvily accommodate setup errors of target and normal tissues for prostate cancer treatment. This work further develops COP to account for (1) deformable organ motion and (2) delineation uncertainties for high-risk prostate cancer patients. The clinical value of COP is investigated by comparing with two margin-based planning techniques: (i) optimized margin (OM) technique that iteratively modifies PTV margins according to the evaluated target coverage probability and (ii) fixed margin (FM) technique that uses empirically selected constant PTV margins. Without patient-specific coverage probability estimation, FM plans are always less immune to the degraded effect of the modeled GUs than the COP plans or the OM plans. Empirical PTV margins face more risks of undesirable target coverage probability and/or excessive dose to surrounding OAR. The value of COP relative to OM varies with different GUs. As implemented for deformable organ motions, COP has limited clinical benefit. Due to optimization tradeoffs, COP often results in target coverage probability below the prescribed value while OM achieves better target coverage with comparable normal tissue dose. For delineation uncertainties, the clinical value of COP is potentially significant. Compared to OM, COP successfully maintains acceptable target coverage probability by exploiting the slack of normal tissue dose in low dose regions and maximally limiting high dose to normal tissue within tolerance.

Prostate cancer is the second-most diagnosed and fatal carcinoma for males in the United States, and better diagnostic markers and potential therapies are needed. microRNAs are small, single-stranded RNA molecules that affect protein expression at the translational level, and dysregulation can dramatically affect cell metabolism. Comparison of 736 microRNA expression levels between the poorly metastatic SV40T immortalized prostate epithelial cell line P69 to its highly tumorigenic and metastatic subline M12 identified 231 miRs that were overexpressed and 150 miRs that showed loss of expression in the M12 cell line. Further evaluation of fourteen identified miRs was accomplished using other prostate cell lines as well as laser-capture microdissected prostate samples. Inhibition of miR-147b was found to affect proliferative, migratory and invasive capabilities of M12 cells, and reduced tumour growth in nude athymic mice. AATF, an activator of the cell-cycle inhibitor p21, was identified as a target. Overexpression of miR-9 was found to affect the epithelial to mesenchymal transition through suppression of e-cadherin, a protein characterized as lost in EMT, as well as suppression of SOCS5, an attenuator of JAK-STAT signaling. Inhibition of miR-9 resulted in reduction of migratory and invasive potential, and significant reduction of tumorigenesis and metastases in male nude athymic mice. miR-17-3p was previously identified as down-regulated in prostate cancer and loss of miR-17-3p shown to cause vimentin transcriptional activation. Reverse phase microarray analysis (RPMA) identified c-KIT as a potential second mRNA target for miR-17-3p. miR-17-3p was shown to modulate not only protein levels, but also messenger RNA levels of c-KIT. Four miR-17-3p binding sites in the c-KIT mRNA were identified. Thus, a number of microRNAs involved in prostate cancer were identified, and their targets found to be highly relevant to tumour progression and could potentially be used as targets for therapy or diagnostics. Stability of microRNAs in forensically relevant biological fluids was evaluated through heat treatment, ultraviolet radiation, and chemical treatment. The dried body fluids showed some susceptibility to harsh treatment, but in most cases microRNAs were still detectable in the samples. microRNAs could represent a highly stable species for body fluid identification methods in forensic science.

Our laboratory is primarily interested in novel pharmacological intervention of cell proliferation and survival pathways expressed in various types of cancer. These cyto-protective pathways can be activated in response to growth factor stimulation, toxic insult and radiation. In our studies, we utilized novel drug combinations with and without radiation to enhance breast & prostate tumor cell death both in vitro and in vivo. Previous studies from our group have shown that UCN-01 and MEK1/2 inhibitors interact to cause tumor cell death in transformed cell lines in vitro. We extended this observation to an in vivo animal model system using the estrogen dependent breast cell carcinoma line MCF-7 and the estrogen independent breast cell carcinoma line MDA-MB-231. This drug combination was shown to profoundly reduce tumor cell proliferation in vivo and also exhibited the ability to significantly reduce ex-vivo tumor cell colony formation 30 days after cessation of the combination drug treatment. In addition, tumor cell death coincided with decreased ERK112 phosphorylation, reduced immunoreactivity of Ki67 and CD31. Overall, these studies demonstrate that UCN-01 and MEK112 inhibitors have the potential to suppress mammary tumor growth in vivo which is independent of p53 status, estrogen dependency, caspase-3 levels or oncogenic K-RAS expression. In our LnCap prostate carcinoma cell studies we demonstrated the impact of hCG and lovastatin in combination with ionizing radiation to radiosensitize and enhance tumor cell lethality. This enhancement was attributed to the hCG-induced activation of ERBB1 via a GPCR, MEK112 and metalloprotease dependent paracrine mechanism which was further enhanced by radiation. This enhanced cell killing effect was shown to involve prolonged activation of PARP1 which could be suppressed by inhibition of ERBB1, MEKl , PI3 kinase or PARP1. Therefore, the combination of hCG, lovastatin and radiation may represent a novel approach to kill prostate cancer cells and potential new therapy.

Previously, the CCR5 receptor was found to be a good target for treating prostate cancer (PCa). Dr. Yan Zhang’s laboratory designed several CCR5 antagonists, which were screened for their inhibitory effect on the growth and invasion of the M12, DU145 and PC-3 PCa cell lines. Primary in vitro screening showed one compound (Drug 17) significantly inhibited the proliferation of PCa cells at 1μM concentration, with a half-maximal inhibitory concentration of 237.68 nM. Further in vitro assays including a proliferation, cytotoxicity and invasion assay confirmed the inhibitory effect of drug 17. The physiological effect of drug 17 was tested by the Ware laboratory in vivo by subcutaneous injection of M12 cells into male, athymic nude mice. Tumor growth was slowed in mice receiving injections of drug 17 compared to sham injected controls. Thus, in vitro and in vivo assays suggest drug 17 might be an effective therapy to block PCa progression.

Pas de résumé français / Pas de résumé anglais

Problématique : Les paramètres fonctionnels extraits en Tomographie par Émission de Positons (TEP) à la 18F-Choline (FCH) et en Imagerie par Résonance Magnétique (IRM) apportent-ils une information supplémentaire par rapports aux informations déjà connues pour caractériser l’agressivité du tissu tumoral ? Objectifs : Notre travail avait pour objectifs tout d’abord (i) de mettre en évidence un éventuel lien entre les paramètres quantitatifs extraits par TEP à la FCH et et les paramètres clinicopathologiques dans le cancer de la prostate. Après cette étude préliminaire, pour quantifier au mieux l’influx de FCH par analyse dynamique de l’acquisition précoce, la deuxième étape avait pour but (ii) d’optimiser cette acquisition. Enfin (iii), le dernier objectif était de mettre en évidence un éventuel lien entre les paramètres quantitatifs extraits par TEP à la FCH et les paramètres quantitatifs extraits par IRM multiparamétrique. Résultats : Pour la première étape (i), nous avons comparé les paramètres quantitatifs extraits par TEP à la FCH et les paramètres clinico-pathologiques de 61 patients venant en TEP à la FCH pour le bilan d’extension initial. L’influx de FCH mesuré par analyse dynamique était plus élevé pour les patients avec un score de Gleason ≥ 4+3 que pour les patients avec un score de Gleason < 4 + 3. Pour la deuxième étape (ii), nous avons tout d’abord voulu optimiser la durée de l’acquisition précoce de TEP à la FCH en comparant le contraste sur bruit de 77 lésions tumorales à 5 minutes et 10 minutes après l’injection de FCH. Le contraste sur bruit à 5 minutes n’était statistiquement pas différent de celui à 10 minutes. La deuxième phase de l’optimisation de l’acquisition précoce de la TEP à la FCH consistait à déterminer quel était le meilleur échantillonnage temporel de ces 5 minutes par la comparaison de 7 échantillonnages temporels différents avec un objectif d’influx de FCH extrait de 37 lésions. L’échantillonnage temporel 12x5”-8x30” a été retenu. Pour la dernière étape (iii), nous avons comparé les paramètres TEP et IRM extraits de 14 lésions tumorales prostatiques. L’influx de FCH et la constante de transfert du gadolinium étaient corrélés, toutefois de manière modérée (r = 0,55). Conclusion : L’influx de FCH mesuré par analyse dynamique de l’acquisition précoce en TEP semble lié à la différenciation des cancers prostatiques. Cet influx semble également lié à la constante de transfert du gadolinium. Ces 2 paramètres d’imagerie semblent toutefois quantifier des processus physiopathologiques différents. Les différents résultats obtenus justifient la poursuite des travaux pour évaluer le rôle de marqueur d’agressivité des cellules cancéreuses prostatiques des différents paramètres quantitatifs obtenus par imagerie fonctionnelle par TEP à la FCH et par IRM multiparamétrique. / Research question: Do the functional parameters derived by 18F-Choline (FCH) Positon Emission Tomography (PET) and Magnetic Resonance Imaging (MRI) add informations to characterize the aggressiveness of prostate cancer? Objectives: The first objective of this work was (i) to enhance a potential link between quantitative parameters extracted by FCH PET and clinicopathological parameters in prostate cancer. Then, after this preliminary study, in order to optimize the quantification of the FCH influx with a kinetic analysis, the second objective was (ii) to optimize the exam protocol of the FCH PET early acquisition. Finally, the last objective was (iii) to enhance a potential link between quantitative parameters extracted by FCH PET and quantitative parameters extracted by multiparametric MRI in prostate cancer. Results: For the first step (i), we compared FCH PET quantitative parameters and clinicopathological parameters extracted from 61 patients referred to the nuclear medicine department to perform an FCH PET/ CT with newly histologically proven prostate cancer and before any treatment The FCH influx measured using kinetic analysis was higher for patients with a Gleason score ≥ 4+3 than for patients with a Gleason score < 4 + 3. Concerning the second step (ii), firstly, we compared the contrast to noise ratio of 77 prostatic cancer lesions at 5 minutes and 10 minutes post-injection in order to optimize the length of the early FCH PET acquisition. No significant difference was observed. Secondly, we sought to define an optimal time sampling of the early FCH PET acquisition comparing 7 different time samplings with a FCH influx as objective extracted from 37 prostatic cancer lesions. The 12x5”-8x30” time sampling was selected. For the last step of this work (iii), we compared FCH PET and multiparametric MRI quantitative parameters extracted from 14 prostatic cancer lesions. The FCH influx was moderately correlated to the vessel permeability measured by the volume transfert constant of gadolinium (r = 0.55). Conclusion: The FCH influx extracted from the early FCH PET acquisition using kinetic analysis seems to be linked to the tumoral differentiation of prostatic cancers. This FCH influx seems also linked to the vessel permeability. However, due to the moderate degree of correlation, these two imaging parameters reflect two different processes. To confirm the results obtained in this work, other studies are needed to enhance the role of the functional parameters derived by FCH PET and multiparametric MRI as biomarkers for prostate cancer.

Le cancer de la prostate est le cancer le plus fréquent chez les hommes. Son développement entraine une néo-angiogénèse qui modifie le réseau capillaire. Il est reconnu que l'IRM dynamique (DCE-MRI) est capable de distinguer ces modifications de la microcirculation physiologique. Cependant, ces images restent difficiles à analyser et à interpréter en routine clinique. Dans cette thèse, nous nous sommes intéressés à la mise en place de méthodes robustes pour l'analyse de ces images. Dans un premier temps, nous traitons les méthodes de quantifications des paramètres pharmacocinétiques. Ainsi, une plateforme logicielle a été construite autour du modèle multi-étapes de Tofts. La validation technique a été conduite en utilisant des images simulées avec connaissance de la vérité de terrain de la distribution des lésions. La validation clinique est en cours dans le service de Radiologie de l'Hôpital Claude Huriez du CHRU de Lille. Parallèlement, nous avons exploré l'application des techniques de traitement des données pour l'analyse non paramétrique et non supervisée des courbes temps-intensités. Nous avons développé une approche originale basée sur la classification spectrale. Cette méthode, basée sur la théorie des graphes, permet le regroupement des signaux après transformation de l'espace de représentation. Par la suite, ces groupes de données peuvent être étiquetés par comparaison avec un signal artériel qui sert de référence. Les expérimentations préliminaires conduites sur les données simulées ainsi que sur des données cliniques montre la faisabilité de l'approche. Les deux approches développées sont complémentaires, l'une donnant des paramètres quantitatifs et l'autre permettant de segmenter les zones cancéreuses. / Prostate cancer is the most common cancer among men. Its developments leads to a neo-angiogenesis that changes the capillary network. It is recognized that the DCE-MRI is able to distinguish these physiological changes in microcirculation. However, the images are difficult to analyze and interpret. In this thesis, we were interested by the development of robust methods for the analysis of these images. Initially, we were focused on pharmacokinetic parameters quantification methods. A software platform was constructed to implement the multi-step Tofts model. Technical validation was performed using simulated images with knowledge of the ground truth. Clinical validation is in progress in the Radiology department of Lille University Hospital. In parallel, we have explored the application of nonparametric and unsupervised techniques of data processing for time-intensity curve analysis. We have developed an original approach based on spectral classification. This method, based on graph theory, allows the grouping of signals after transformation of the space of representation. Subsequently, these groups of data can be labeled by comparison to the arterial signal serving as reference. Preliminary experiments conducted on simulated data as well as clinical data show the feasibility of the approach. The two approaches are complementary, one giving quantitative parameters and the other segmenting the cancerous areas.