Philosophiae Doctor - PhD / Introduction: Globally, there are over 190,000 new reported cases of ovarian cancers per annum. This comprises 3% to 4% of all cancers in women. Ovarian cancer is one of the leading causes of deaths in women. Ovarian cancer is the second most diagnosed gynaecological malignancy and over all the fifth cause leading to death among all types of cancer in the UK in 2004. More than 70% of epithelial ovarian cancers are diagnosed at an advanced stage. Consequently, the prognosis is poor and the mortality rate high. Thus, the survival rate is affected by how far the disease has progressed or spread. A dire need exists to identify ovarian cancer biomarkers, which could be used as good indicators of expression in ovarian cancer cells in vitro Aim: The aim of this study was to analyse selected cancer biomarkers, which are currently under intense investigation for their suitability to diagnose epithelial ovarian cancer at an early stage. These biomarkers were analysed in terms of their in vitro expression in normal epithelial cells and ovarian cancer cell lines, which allows for their genomic and proteomic classification. The expression analysis of each biomarker is related to the malignancy of a tumour and, therefore, advocates its use for potential future improvement of sensitive tumour markers. Methods: The primary human ovarian surface epithelial cell line (HOSEpiC), SKOV-3 cells and the OAW42 human epithelial ovarian tumour cell lines were used to evaluate the selected cancer biomarkers. Cells were cultured using appropriate media and supplements, and real-time quantitative polymerase chain reaction (RT-PCR) utilized to validate expression levels of the following genes: HDAC1, HDAC2, HDCA3, HDAC5, HDAC6, HDAC7, HDAC8, LPAR1, LPAR2, MUC16 and FOSL1, against normal housekeeping genes GAPDH and HPRT. In addition, immunocytochemistry was also used in the validation process of the aforementioned genes. Significance: ovarian cancer cells express gene signatures, which pose significant challenges for cancer drug development, therapeutics, prevention and management. The present study is an effort to explore ovarian cancer biomarkers to provide a better diagnostic method that may offer translational therapeutic possibilities to increase five- year survival rate. Results: HDAC5, HDAC6, LPAR1, LPAR2 and MUC16 expressed distinctively in ovarian cancers matched to other tissues or cancer types have already been identified by RT-QPCR and confirmed by immunocytochemistry and efforts to generate monoclonal antibodies to the other six genes (HDAC1, HDAC2, HDAC3, HDAC7, HDAC8 and FOSL1) encoded proteins are underway. Conclusions: here we provide strong evidence suggesting that HDAC5, HDAC6, LPAR1, LPAR2, except MUC16 are up regulated in ovarian cancer. These data were confirmed by examining Human Protein Atlas (HPA) databases, in addition to protein expression of HDAC5, HDAC6, LPAR1, LPAR2 and MUC16 in cells cytoplasm. For future prospective, using other techniques that assess the variant expression that could explain the release of these gene candidates into the circulation with serum tumour markers, and protein expression will be strengthened.

Les cancers de l’ovaire présentent un taux de survie à 5 ans inférieur à 40% et constituent la principale cause de décès par cancer gynécologique dans le monde. Ce sombre pronostic s’explique par un diagnostique tardif (du à un développement asymptomatique dans les premiers stades) et une résistance aux traitements existants et souligne la nécessité de développer de nouvelles approches thérapeutiques. La découverte ces dernières années, d’un nombre important de lncRNAs a permis d’ouvrir de nouvelles perspectives pour la recherche en oncologie. Peu d’études ont à ce jour exploré leur implication dans la chimiorésistance, a fortiori dans les cancers de l’ovaire. Parmi ces lncRNAs, ‘UCA1’ exerce de multiples fonctions oncogéniques par des mécanismes encore peu décrits. Son expression constitue un facteur de mauvais pronostique dans diverses tumeurs malignes dont les cancers de l’ovaire. Nous avons pu démontrer un rôle de ceRNA pour le miR-27a-5p, qui une fois libéré suite à l’inhibition d’UCA1, régule négativement UBE2N, une cible directe. UBE2N est un acteur connu des voies de réparation de l’ADN et de la régulation de la voie NF-kB, et son inhibition dans nos modèles entraîne une augmentation de l’expression de BIM et perturbe les mécanismes de réparation de l’ADN, sensibilisant des cellules cancéreuses ovariennes à l’action des sels de platine et des PARPi. L’inhibition d’UBE2N sensibilise également aux sels de platine plusieurs cultures organoïdes tridimensionnelles dérivées de patientes, et pourrait ainsi constituer une stratégie thérapeutique innovante pour lutter contre la chimiorésistance dans le cancer de l'ovaire. / Ovarian cancers present a 5-year survival rate of under 40% and are the leading cause of death from gynecological cancer worldwide. This poor prognosis is explained by a late diagnosis (due to asymptomatic development in the early stages) and resistance to existing treatments and highlights the need to develop new therapeutic approaches. The discovery in recent years of a significant number of lncRNAs has opened up new opportunities for oncology research. Few studies have so far explored their involvement in chemoresistance, let alone ovarian cancer. Among these lncRNAs, 'UCA1' performs multiple oncogenic functions through mechanisms not yet well described. Its expression is a factor of poor prognosis in various malignant tumours including ovarian cancers. We were able to demonstrate a role of ceRNA for miR-27a-5p, which when released following the inhibition of UCA1, negatively regulates UBE2N, a direct target. UBE2N is a known actor in DNA repair pathways and NF-kB pathway regulation, and its inhibition in our models leads to an increase in BIM expression and disrupts DNA repair mechanisms, sensitizing ovarian cancer cells to the action of platinum salts and PARPi. The inhibition of UBE2N also sensitizes to platinum salts, several three-dimensional organoid cultures derived from patients, thus may provide an innovative therapeutic strategy to combat chemoresistance in ovarian cancer.

Les cancers épithéliaux de l’ovaire sont la première cause de décès par cancer gynécologique. Dans ce travail, nous nous sommes intéressés aux cancers ovariens séreux de haut grade (HGSOC) et de stade avancé, pour lesquels les options thérapeutiques demeurent limitées. Plusieurs laboratoires, dont le nôtre, ont identifié des groupes moléculaires distincts au sein de ces HGSOC sur la base de données transcriptomiques.Dans toutes ces études, un sous-groupe moléculaire, nommé « fibrose » ou« mésenchymateux », est systématiquement observé et invariablement associé à un pronostic sombre pour les patientes. La signature transcriptomique qui identifie ce groupe de patientes inclue de nombreux gènes impliqués dans le remodelage de la matrice extracellulaire et la composition stromale, suggérant un rôle possible du stroma dans l’agressivité de ce sous-groupe moléculaire particulier d’HGSOC.Par une étude combinant de façon originale l’analyse concomitante de 6 marqueurs de fibroblastes associés au cancer (CAF), nous mettons en évidence pour la première fois l’existence de 4 sous-populations de CAF. De façon intéressante, l’une de ces 4 populations(dite CAF-S1) s’accumule significativement dans le sous-type moléculaire« fibrose/mésenchymateux » des HGSOC. Or, cette population particulière de CAF-S1 présente une activité immunosuppressive, en favorisant non seulement la survie ou l’attraction des lymphocytes T régulateurs, mais également en stimulant leur activation par l’expression du facteur de transcription FOXP3 (Forkhead Box P3). Enfin, nous identifions l’isoforme β de CXCL12 (chimiokine à motif C-X-C ligand 12) comme un acteur majeur de l’identité et de la fonctionnalité de ces CAF-S1 immunosuppresseurs. En effet, CXCL12βs’accumule spécifiquement dans les CAF-S1 et joue un rôle clé dans la fonction immunosuppressive de ces CAF au sein du stroma des HGSOC mésenchymateux.Nos résultats mettent ainsi en évidence un mécanisme expliquant, au moins en partie, le pronostic sombre des patientes atteintes d’HGSOC mésenchymateux. La caractérisation approfondie du stroma de ces tumeurs agressives permet d’envisager de nouvelles stratégies thérapeutiques ciblant à la fois les CAF et les cellules immunitaires dans le but d’améliorer la survie des patientes mésenchymateuses. / Epithelial ovarian cancers are the first cause of death from gynecologic cancer. We focusedour work on high-grade serous ovarian cancer (HGSOC) patients, for who only very fewtherapeutic options exist. In the past, several laboratories, including ours, have identified -based on transcriptomic data- distinct molecular subgroups amongst HGSOC. Interestingly,among these different molecular subgroups, one of them, referred to as “Fibrosis” or“Mesenchymal” is systematically identified and consistently associated with poor patientprognosis in all studies. Transcriptomic signature that defines this specific molecularsubgroup of HGSOC contains mainly genes involved in extracellular matrix remodeling andstromal composition, suggesting a potential role of stroma and Carcinoma-AssociatedFibroblasts (CAF) in this particular “Fibrosis/Mesenchymal” HGSOC subgroup.By combining various technics studying concomitantly 6 different CAF markers, we identifiedfor the first time 4 different subpopulations of CAF in HGSOC. Interestingly, one of thesesub-populations, referred to as CAF-S1, significantly accumulates in the“Mesenchymal/Fibrosis” subgroup of HGSOC. Importantly, we demonstrated that the CAFS1cellular subpopulation exhibits immunosuppressive activities. Indeed, CAF-S1 fibroblastsnot only by attract regulatory T lymphocytes but also promote their survival and activation(assessed by expression of FOXP3). Finally, we uncovered the specific role of the CXCL12βisoform as an important player of CAF-S1 identity and immunosuppressive functions inmesenchymal HGSOC.All together, these results identify a stromal heterogeneity in HGSOC, which has beenbroadly underestimated until now. Moreover, our work demonstrates the accumulation of aCAF subpopulation with immunosuppressive functions in HGSOC mesenchymal patients thatcould account, at least in part, for their poor survival rate. Deep characterization of thestroma may enable us to define new therapeutic options combining CAF-targeting therapiesand immunotherapies, in order to improve survival of HGSOC mesenchymal patients.

The layer of cells lining the outer surface of the mammalian ovary, the ovarian surface epithelium (OSE), is a constant feature throughout the dynamic tissue remodeling that occurs throughout the reproductive cycle (follicle growth, ovulation, corpora lutea formation and pregnancy). Abnormal development of these cells is responsible for 90% of all epithelial ovarian cancers in women and epidemiological studies have shown that susceptibility to ovarian cancer is negatively correlated with increasing pregnancy. Little is known about how OSE cells are affected at each stage of the cycle, so the main aim of this study was to determine how the reproductive cycle affected proliferation and degeneration of OSE cells. This study utilised three animal models each with a different type of reproductive cycle: a mono-ovular seasonal breeder (Sheep), a mono-ovular polyoestrous breeder (Cow) and a poly-ovular non human primate (marmoset) to allow comparisons to be made. Comparison of OSE proliferative activity was made in sheep and marmoset at each stage of the cycle including pregnancy and anoestrous. The bovine model was used to investigate apoptotic cell death. Proliferative activity of somatic cells within the sheep ovary was monitored throughout the reproductive cycle by detection of cell cycle markers PCNA and Ki67 using immunohistochemistry. The pattern of OSE proliferation was correlated with the pattern of follicle development at each stage (sheep and marmoset). During pregnancy cell proliferation was significantly lower in OSE and in granulosa cells, reflecting a suppression of mature follicle development during these stages whereas in cycling animals proliferation was increased. Differences in OSE proliferation were observed in relation to the local underlying tissue environment in both sheep and marmoset. Epithelial cell rupture and regeneration enhanced the hormonal mitogenic action on epithelial cells, which showed highest proliferation over corpora lutea in each animal model. To test the hypothesis that these changes are mediated by hormones or growth factors ovine OSE cells were cultured and proliferative activity monitored after treatment with several factors: fetal calf serum (FCS), follicular fluid from follicles of varying sizes, corpora lutea extracts, recombinant human IGF-1, oestradiol and progesterone. IGF alone was demonstrated to have an affect on increasing proliferation of cultured OSE cells. Levels of FSHr and LHr were monitored by quantitative real- time PCR and it was demonstrated that the concentration of gonadotrophin receptors in OSE, increased prior to and after ovulation, at which time the in vivo OSE proliferation also peaked. The in situ apoptosis index was determined in bovine tissue using TUNEL throughout the regular cycle, and at mid and late-pregnancy stages. The results showed that pregnancy induced apoptotic activity in OSE cells and up regulated the tumour suppressor gene p53. Cultured bovine OSE cells also exhibited an increased level of apoptosis following progesterone treatment. Since p53/p53 gene expression in OSE over the corpora lutea producing progesterone also increased, this progesterone-mediated apoptosis may be mediated through an up-regulation of p53 synthesis. The effect of pregnancy and low production of gonadotrophins in the regulation of OSE cell morphology and activity was further investigated in the marmoset monkey (a non-human primate) treated with GnRH antagonist and infused with BrdU to monitor proliferative activity. OSE proliferation was correlated to ovarian events (follicular growth, ovulation and luteinization) and this was suppressed during pregnancy. Inhibition of gonadotrophin secretion by treatment with a GnRH antagonist also markedly inhibited OSE proliferation. Taken together these studies support the hypothesis that pregnancy and periods of anovulation reduce proliferation of OSE cells and alter the pattern of apoptotic cell death and that this effect is independent of species and reproductive pattern. Suppression of gonadotrophins and other growth factors during pregnancy could enhance p53-mediated apoptosis of damaged and mitogenic cells arising from repeated ovulations. This effect may partly explain why increasing number of pregnancies in woman reduces the chance of epithelial ovarian cancers.

Les cancers ovariens se distinguent par leur faible incidence, associée à une mortalité élevée, et représente la cinquième cause de mortalité par cancer pour la population féminine. Cette mortalité est due principalement à l’absence de symptômes aux stades précoces des cancers ovariens, retardant leur diagnostic, majoritairement posé lorsqu’une carcinomatose péritonéale est installée. La cavité abdominale est alors envahie par de nombreuses métastases. Le traitement de référence est la chirurgie de cytoréduction complète et la chimiothérapie par voie intraveineuse. Lors de la chirurgie, le chirurgien doit inspecter la totalité de la surface de la cavité péritonéale, et éliminer des tumeurs de toutes tailles, mêmes submillimétriques. Faute de pouvoir détecter toutes les métastases, la cytoréduction est souvent partielle, avec pour conséquence une diminution des chances de guérison. Afin de guider le chirurgien et améliorer le pourcentage de cytoréduction complète, des sondes fluorescentes, conçues pour marquer spécifiquement les tumeurs, ont été développées. Malgré des résultats encourageants, ces sondes souffrent de nombreuses limitations en termes de brillance, de stabilité, et de modularité. Dans ce contexte, de nombreuses nanoparticules, capables de passer outre ces limitations, suscitent un grand intérêt. Parmi celles-ci, des nanocristaux semi-conducteurs, appelés quantum dots, se distinguent par une brillance exceptionnelle. Notre étude s’est basée sur ces quantum dots, associés à un agent de ciblage de référence pour les cancers ovariens, le folate. D’abord testées sur modèles cellulaires et tumoraux in vitro, ces nanoparticules ont démontré de bonnes capacités d’imagerie. Ces propriétés ont ensuite été expérimentées sur modèle murin de carcinomatose péritonéale. Enfin, la bioaccumulation des quantum dots restant un obstacle à leur application clinique, différentes alternatives ont été appliquées pour tenter d’obtenir leur excrétion / Ovarian cancers have a low incidence but a high mortality rate, making them the fifth cause of death by cancer for female population. This high mortality rate is associated with the absence of symptom at the early stage of ovarian cancer, often delaying the diagnosis to advanced stages, mainly peritoneal carcinomatosis. At this stage, metastases have already invaded the abdominal cavity. The reference treatment combines a complete cytoreduction surgery and intravenous chemotherapy. During the cytoreduction, the surgeon must inspect the whole peritoneal surface, and remove all of all sizes, even sub-millimetric. Because of the difficulty to detect and to remove every cancerous tissue, cytoreduction is frequently incomplete, thus reducing the recovery rate. To guide the surgeon and to increase the percentage of complete cytoreduction, fluorescent probes have been developed to target tumors specifically. Despite encouraging results, these probes suffer from many limitations such as low brightness, weak stability and cumbersome modularity. In this context, nanoparticles, that are able to outpass these limitations, have generated a growing interest. Among these nanoparticles, semiconductor nanocristals, called quantum dots, display an exceptional brightness. We investigated these quantum dots, associated with folate, a reference targeting agent for ovarian cancers. Firstly investigated on in vitro cellular and tumor model, folate targeted nanoparticles show encouraging imaging capabilities. These capabilites were also experimented on peritoneal carcinomatosis murine model. Finally, the main obstacle to quantum dot clinical application remains their bioaccumulation, therefore, different alternatives were explored to achieve excretion

BRCA1 and BRCA2 are gene mutations that drastically increase chances of developing breast and ovarian cancers, up to 20-fold, for women. A genetic blood test is used to detect BRCA mutations. Though these mutations occur in one of every 400 in the general population (excluding Ashkenazi Jewish ethnicity), they are present in most cases of hereditary breast and ovarian cancer patients. Hence, it is common practice for the physicians to require genetic testing for those that fit the rules as recommended by the National Cancer Comprehensive Network. However, data from the Myriad Laboratory, the only provider of the test until 2013, show that over 70 percent of those tested are negative for BRCA mutations [1]. As there are significant costs and psychological trauma associated with having to go through the test, there is a need for more comprehensive rules for determining who should be tested. Once the presence of BRCA is identified via testing, the next challenge for both mutation carriers and their physicians is to select the most appropriate types and timing of intervention actions. Organizations such as the American Cancer Society suggest drastic intervention actions such as prophylactic surgeries and intense breast screenings. These actions vary significantly in their cost, cancer incidence prevention ability, and can have major side effects potentially resulting in reproduction inability or death. Effectiveness of these intervention actions is also age dependent. In this dissertation, both an analytical and an optimization framework are presented. The analytical framework uses supervised machine learning models on extended family history of cancers, and personal and medical information from a recent nationwide survey study of women who have been referred for genetic testing for the presence of a BRCA mutation. This framework provides the potential mutation carriers as well as their physician with an estimate of the likelihood of having the mutations. The optimization framework uses a Markov decision process (MDP) model to find cost-optimal and/or quality-adjusted life years (QALYs) optimal intervention strategies for those tested positive for a BRCA mutation. This framework uses a dynamic approach to address this problem. The decisions are made more robust by considering the variation in estimates of the transition probabilities by using a robust version of the MDP model. This research study delivers an innovative decision support tool that enables physicians and genetic consultants predict the population at high risk of breast and ovarian cancers more accurately. For those identified with presence of the BRCA mutation, the decision support tool offers effective intervention strategies considering either minimizing cost or maximizing QALYs to prevent incidence of cancers.

<p> The ErbB2/Neu orphan receptor tyrosine kinase is amplified in 20-30% of breast and ovarian cancers and predicates poor patient prognosis. Five conserved tyrosine residues, autophosphorylated by Neu catalytic activity, "dock" adaptors and second messengers that activate discrete signalling pathways, most prominently the Ras/MAPK pathway, to regulate cell survival and proliferation. </p> <p> Genetic analysis using Drosophila provides an efficient means for identifying evolutionarily conserved signalling components. Neu and Drosophila EGFR overexpression directs Drosophila tissue development synonymously. Consistent with biochemical evidence, genetic analysis ofNeu signalling through individual pTyr revealed activating signals for 4(Yl144, YB; Y1201, YC; Y1226/7, YD; Y1253, YE) of the 5 sites. Strong Ras-dependent signalling was mediated through adaptors Grb-2 (YB) and SHC (YD). In contrast to biochemical evidence, a strong Ras component was not genetically detected for YC or YE. </p> <p> We have conducted two enhancer-suppressor screens to identify novel Ras and non-Ras requirements for YC signalling. For the first screen, a quantitative approach was designed to identify modification of an YC-specific wing notch phenotype. Thirty-two members of the Ras/MAPK signalling cassette were assessed. Sensitivity to Ras, Raf, MAPK and the Ras-related GTPase R-Ras was identified downstream ofYC. The second screen, a large-scale mutagenesis, took advantage of an YC-induced rough eye phenotype. From over 19 000 screened flies, 11 enhancers and 6 suppressors were isolated. One strong suppressor has been genetically mapped to the dual transcription factor and phosphatase eyes absent. Other promising Ras-dependent and Ras-independent modifiers await farther mapping. Results from both screens point to YC as a unique pTyr that uses both Ras-dependent and Ras-independent outputs. </p> / Thesis / Doctor of Philosophy (PhD)