On constate que la majorité des essais de phase III, conduits après des essais de phase II pourtant prometteurs, sont “négatifs”, la nouvelle thérapeutique se révélant finalement trop toxique ou insuffisamment efficace. L’hétérogénéité de la population participant aux différentes phases de développement est une explication. Elle induirait une estimation erronée de la toxicité et, par dilution de l’effet traitement, conduirait à arrêter l’évaluation thérapeutique alors que peut être un sous-ensemble de cette population, définie à partir d’une caractéristique particulière, pourrait en bénéficier.Dans cette thèse, nous proposons dans un premier temps une réflexion sur les aspects méthodologiques des essais de phase II qui permettraient d’améliorer l’identification précoce des thérapeutiques toxiques et des populations les plus sensibles et donc de ne planifier des essais de phase III que sur des populations encore mieux ciblées. Dans un second temps, nous présentons une nouvelle méthodologie d’essai de phase II que nous avons développée pour prendre en compte l’hétérogénéité de la population et son intérêt en pratique clinique courante. Avec cette méthode, qui est une extension du plan de Fleming à deux étapes, le développement des médicaments est moins fréquemment arrêté pour la population entière et moins de patients non sensibles à la nouvelle thérapeutique sont exposés à des molécules potentiellement toxiques, durant l’étape 2 de l’essai de phase II ou plus tard lors de l’essai de phase III. / The majority of phase III clinical trials, despite being conducted after promising phase II trials, are "negative," with the new therapy determined in the end to be too toxic or insufficiently efficacious. One explanation is the heterogeneity of the populations participating in various phases of development, which results in an erroneous estimation of the toxicity and thus a diluted therapeutic effect. This may lead to termination of evaluation of a therapy, even if a sub-population, defined by a particular characteristic, may stand to benefit from it. In this thesis, we propose a close examination of the methodological aspects of phase II trials which would permit improved early identification of toxic therapies and of responsive populations, so that phase III trials may be designed only with the best targeted populations in mind. We present as well a new phase II clinical trial methodology which we have developed to take into account trial population heterogeneity and its importance in current clinical practice. With this method, drug development is less often stopped for the entire phase II population and less non sensitive patients are exposed to toxic drugs in the second part of phase II trials, and next in phase III trials.

Das ITC SFN und der Mikronährstoff Se sind bekannt als chemopräventive Inhaltsstoffe von Gemüse der Brassica-Familie, welcher auch Brokkoli angehört. Die Wirkungen von sowohl SFN als auch Se beruhen auf zahlreichen verschiedenen Mechanismen. Es existieren jedoch Schnittstellen, an welchen Interaktionen beider Substanzen möglich sind. Basierend auf diesem Wissen wurden in dieser Arbeit Wechselwirkungen zwischen SFN und Se auf die Aktivität sowie Expression von Phase II Enzymen und Selenoproteinen untersucht. Der Einfluss der Kombination von SFN und Se auf die unter physiologischen Bedingungen stattfindende Proliferation und Apoptose war ebenso Gegenstand der Arbeit wie die Modulation von Entzündungsprozessen sowie der Tumorentstehung während der entzündungsverstärkten Colonkanzerogenese im Mausmodell. Das hinsichtlich seiner Wirksamkeit mit aus GRA hydrolysiertem SFN zunächst als vergleichbar befundene synthetische SFN wurde für die Untersuchung im AOM/DSS-induzierten Colontumormodell gewählt und in Kombination mit 3 verschiedenen Selendiäten verabreicht. Der Einfluss von SFN und Se auf Phase II Enzyme und Selenoproteine entlang des GIT war organabhängig und nach 4 Wochen geringer als nach 7 Tagen. Die schwächere Induktion deutet auf eine Anpassung des Organismus hin. Ein SFN-vermittelter Effekt auf NQO1 war im Selenmangel am deutlichsten. Die Aktivität des Selenoproteins TrxR wurde hingegen erst bei ausreichender Selenversorgung durch SFN beeinflusst. Die als Nrf2-Zielgen bekannte und in der Hierarchie der Selenoproteine einen hohen Rang einnehmende GPx2 konnte in bestimmten Organen bereits unter selenarmen Bedingungen durch SFN induziert werden. Eine Überexpression des Enzyms war jedoch nicht möglich. SFN steigerte, unabhängig vom Selenstatus, im oberen Abschnitt des GIT und im Colon die Aktivität der GST. Eine Induktion des eigenen Metabolismus wäre somit denkbar. Im Falle eines Mangels an GPx2 wurde GPx1 bei hinreichender Selenversorgung stärker exprimiert, allerdings konnte sie die Funktion von GPx2 nicht völlig erset-zen. Im Selenmangel kann die Aktivitätssteigerung der TrxR im Dünndarm, dem Ab-schnitt der Selenabsorption, als ein Versuch der GPx2-Kompensation angesehen werden. SFN war nicht in der Lage, über eine Aktivierung des Nrf2/ARE-Signalweges kompensatorische Effekte zu induzieren. Apoptotische Prozesse wurden unter physiologischen Bedingungen nur marginal durch SFN und Se moduliert. Das elektrophile ITC konnte lediglich im Selenmangel Apoptose im luminalen Bereich der Colonkrypten induzieren. Die durch supranutritive Selenkonzentration induzierte Apoptose im Kryptengrund wurde nicht durch SFN beeinflusst. Einer bei Abwesenheit der GPx2 erhöhten Apoptoserate im Kryptengrund wirkte SFN bei adäquater Selenversorgung entgegen, war indessen proapoptotisch unter selendefizienten Konditionen. Der Einfluss von SFN auf die Entzündung war deutlich abhängig vom Selenstatus. Während SFN im Selenmangel anscheinend prooxidative Prozesse induzierte und die Entzündungssymptome verschlimmerte, wirkte es unter adäquatem Selenstatus an-tiinflammatorisch. Den vergleichsweise milden Grad der Entzündung im selensupplementierten Status konnte SFN nicht zusätzlich beeinflussen. SFN veränderte die Inzi-denz colorektaler Tumore nicht. Ein, die Tumorinitiation blockierender SFN-Effekt durch direkte Hemmung der metabolischen Aktivierung des Prokanzerogens im selenadäquaten Zustand scheint offensichtlich. Eine Überversorgung mit Se kann protektiv im Hinblick auf Entzündung oder Colonkanzerogenese sein, jedoch bewirkt SFN keinen zusätzlichen Schutz. Kombinationseffekte von SFN und Se in Bezug auf Phase II Enzyme, Selenoproteine und Apoptose sowie die entzündungsverstärkte Colonkanzerogenese sind nicht eindeutiger Natur und können, abhängig vom Endpunkt, synergistische oder antagonistische Züge aufweisen. Eine bei Selendefizienz deutlichere Wirkung von SFN kann mit Hilfe der gesteigerten Aktivierung von Nrf2 erklärt werden, dies muss jedoch nicht von Vorteil sein. Bei adäquater Selenversorgung kann SFN kurzfristig antiinflammatorische und antikanzerogene Prozesse induzieren. Von einer längerfristigen ständigen SFN-Aufnahme in Form von GRA-reichen Brassicacea ist jedoch abzuraten, da von einer Adaptation auszugehen ist. Die Wirkung von SFN innerhalb der komplexen Pflanzenmatrix bleibt Gegenstand zukünftiger Untersuchungen. / Sulforaphane (SFN), a versatile actor derived from broccoli or other brassicaceae, is proposed to be a dietary anticarcinogen. Together with an adequate selenium status, it has been associated with a decreased risk for developing certain forms of cancer. In our mouse model, we investigate the influence of SFN and Se on the expression and activity of selenoproteins and phase II enzymes as well as the effects on inflammation triggered colon carcinogenesis. SFN increased NQO1 activity and protein expression significantly in the ileum, in both, Se-deficiently and Se-adequately fed animals. TrxR activity was increased in Se-adequately compared to Se-deficiently fed mice, SFN positively affected TrxR activity only in the former ones. An increase of GPx2 protein expression by SFN was observed in the ileum of mice of both diets. GPx1 reacts sensitively on Se supply. GST was the only enzyme analyzed being significantly increased by SFN on activity level in the colon. All AOM/DSS treated animals showed an inflammation, which was attenuated by SFN within Se-adequacy. In contrast, Se-deficient animals showed a more severe inflammation. The administration of SFN therefore seemed to enhance this even more and to be not beneficial in this case. SFN inhibited colon carcinogenesis in Se-adequate mice when being administered together with AOM. To summarize, both, GPx2 and TrxR, require selenium in order to be synthesized. In contrast to TrxR, the SFN-mediated induction of GPx2, the highest ranking selenoprotein, does not depend on additional selenium supply. Whereas distinct effects by SFN were observed in the ileum, only GST was influenced by SFN in the colon. SFN seems to induce its own metabolism. In conclusion, SFN and Se attenuate inflammation and colon carcinogenesis, preferably by means of up-regulating the endogenous defense system and inhibiting the metabolic activation of AOM.

Le cancer demeure en France un problème majeur de santé publique : Cette pathologie reste la principale cause de mortalité, avec près de 150 000 décès chaque année. Afin de lutter contre cette maladie, le plan cancer 2014-2019 a posé comme objectif de développer la recherche clinique, en incluant dès 2019 au moins 50 000 patients par an dans des essais thérapeutiques. La recherche clinique en oncologie est en plein essor depuis l’avènement de thérapies telles que les thérapies moléculaires ciblées ou l’immunothérapie. Le nombre de nouveaux médicaments en développement en oncologie a augmenté de façon exponentielle depuis les années 2000. Il en découle un accroissement important du nombre de publications d’études en oncologie. L’analyse d’une étude n’est possible que si son schéma expérimental, sa réalisation et ses conditions d’analyse sont clairement décrits dans l’article publié. Le reporting est le fait de décrire dans une publication toutes les actions, notamment d’ordre méthodologiques, réalisées au cours de l’étude. 5 revues systématiques de la littérature ont été réalisées afin d’analyser le reporting des articles d’études en oncologie. La première étude a objectivé que la littérature était pauvre en articles analysant le reporting des études observationnelles (2% des publications) et que le reporting intrinsèque des revues systématiques récentes portant sur le sujet était exhaustif (score OQSR moyen (2010-2014) = 10, écart-type =2). 826 articles d’études ont été analysés dans la deuxième revue systématique, dont 84.5% d’études observationnelles. Les reporting des variables et du critère de jugement principal étaient clairs, respectivement dans 100% et 85.2% des cas. Une discordance de reporting pour le test statistique principal utilisé, entre les parties Méthodes et Résultats, a été identifié dans 23 articles. La troisième revue systématique a mis un évidence un manque de données sur le reporting des essais de phase II en oncologie (13 articles) et 3 scores d’évaluation du reporting avaient été décrits : score OQS, score KMS et score d’index de qualité. 557 publications d’essais de phase II ont été étudiées dans la quatrième revue systématique. Le score de reporting KMS était égal à 0, 1, 2, 3 respectivement pour 3,9%, 21,4%, 48,5% et 26,2% articles. En analyse multivariée, les 3 variables associées à la valeur du score KMS étaient : Le reporting clair du design statistique (OR = 2.22; IC 95% [1.36-3.65] ; p <0.001), l’essai monocentrique (OR = 0.25 [0.09 – 0.74] ; p = 0.012) et l’analyse menée per protocole (OR = 0.48; IC 95% [0.32 -0.72] ; p <0.001). La dernière revue systématique a porté sur 557 articles d’essais de phase II. 182 des essais de phase II (32.7%) ont abouti à la réalisation d’un essai de phase III, dont 57 avaient un critère de jugement principal positif et statistiquement significatif. En conclusion, le reporting des études en oncologie reste perfectible. Les points d’optimisation relevés sont l’élaboration de recommandations de reporting spécifique aux essais de phase II en oncologie et l’adhésion de l’ensemble des journaux médicaux aux différentes recommandations de reporting / Cancer remains a major public health problem in France : This pathology is the leading cause of death, with nearly 150,000 deaths each year. In order to fight against this disease, the 2014-2019 French cancer plan set the goal of developing clinical research, including at least 50,000 patients per year in 2019 in therapeutic trials. Clinical research in oncology has been booming since the advent of therapies such as targeted molecular therapies or immunotherapy. The number of new drugs being developed in oncology has increased exponentially since the 2000s, resulting in a significant increase in the number of oncology study publications. The analysis of a study is possible only if its experimental design, its realization and its conditions of analysis are clearly described in the published article. Reporting is the process of describing in a publication all the actions, including methodological ones, carried out during the study. 5 systematic reviews were conducted to analyze the reporting of oncology study articles. The first study showed that the literature was poor in articles analyzing the reporting of observational studies (2% of publications) and that the intrinsic reporting of recent systematic reviews on the subject was exhaustive (average OQSR score (2010-2014) = 10, standard deviation = 2). 826 articles of study were analyzed in the second systematic review, 84.5% of which were observational studies. The reporting of the variables and the primary endpoint were clear in 100% and 85.2% of the cases, respectively. A reporting discrepancy for the main statistical test used, between the Methods and Results parts, was identified in 23 articles. The third systematic review revealed a lack of data on the reporting of phase II trials in oncology (13 articles) and 3 reporting evaluation scores were described: OQS, KMS and quality index score. 557 phase II trial publications were reviewed in the fourth systematic review. The KMS reporting score was 0, 1, 2, 3 respectively for 3.9%, 21.4%, 48.5% and 26.2% publications. In multivariate analysis, the 3 variables associated with the KMS score were : Reporting of a clear statistical design (OR = 2.22, 95% CI [1.36-3.65], p <0.001), the single-center trial (OR = 0.25 [0.09 - 0.74], p = 0.012) and per protocol analysis (OR = 0.48, 95% CI [0.32-0.72], p <0.001). The last systematic review focused on 557 Phase II trial articles. 182 Phase II trials (32.7%) resulted in a Phase III trial, of which 57 had a positive and statistically significant primary endpoint. In conclusion, reporting of studies in oncology remains perfectible. Optimization points identified are the development of specific reporting recommendations for phase II trials in oncology and the adherence of all medical journals to the various reporting recommendations

Advancing applications of engineered nanomaterials (ENM) in various fields create the opportunity for intended (e.g. drug and gene delivery) or unintended (e.g. occupational and environmental) exposure to ENM. However, the knowledge of ENM-toxicity is lagging behind their application development. Understanding the ENM hazard can help us to avoid potential human health problems associated with ENM applications as well as to increase their public acceptance. Ceria (cerium [Ce] oxide) ENM have many current and potential commercial applications. Beyond the traditional use of ceria as an abrasive, the scope of ceria ENM applications now extends into fuel cell manufacturing, diesel fuel additives and for therapeutic intervention as a putative antioxidant. However, the biological effects of ceria ENM exposure have yet to be fully defined. Both pro-and anti-oxidative effects of ceria ENM exposure are repeatedly reported in literature. EPA, NIEHS and OECD organizations have nominated ceria for its toxicological evaluation. All these together gave us the impetus to examine the oxidative stress effects of ceria ENM after systemic administration. Induction of oxidative stress is one of the primary mechanisms of ENM toxicity. Oxidative stress plays an important role in maintaining the redox homeostasis in the biological system. Increased oxidative stress, due to depletion of antioxidant enzymes or molecules and / or due to increased production of reactive oxygen (ROS) or nitrogen (RNS) species may lead to protein oxidation, lipid peroxidation and/or DNA damage. Increased protein oxidation or lipid peroxidation together with antioxidant protein levels and activity can serve as markers of oxidative stress. To investigate the oxidative stress effects and the mechanisms of ceria-ENM toxicity, fully characterized ceria ENM of different sizes (~ 5nm, 15nm, 30nm, 55nm and nanorods) were systematically injected into rats intravenously in separate experiments. Three brain regions (hippocampus, cortex and cerebellum) were harvested from control and ceria treated rats after various exposure periods for oxidative stress assessment. The levels of oxidative stress markers viz. protein carbonyl (PC), 3-nitrotyrosine (3NT), and protein bound 4-hydroxy-2-trans-nonenal (HNE) were evaluated for each treatment in each control and treated rat organ. Further, the levels and activities of antioxidant proteins, such as catalase, glutathione peroxidase (GPx), glutathione reductase (GR), super oxide dismutase (SOD), were measured together with levels of heat shock proteins heme oxygenase -1 and 70 (HO-1 and Hsp-70). In addition, the levels of pro-inflammatory cytokines IL-1β, TNF-α, pro-caspase-3, and autophagy marker LC-3A/B were measured by Western blot technique. In agreement with the literature-proposed model of oxidative stress hierarchy mechanism of ENM-toxicity, the statistical analysis of all the results revealed that the ceria ENM-induced oxidative stress mediated biological response strongly depends on the exposure period and to some extent on the size of ceria ENM. More specifically, a single intravenous injection of ceria ENM induced tier-1 (phase-II antioxidant) response after shorter exposure periods (1 h and 20 h) in rat brain. Upon failure of tier-1 response after longer exposure periods (1 d to 30 d), escalated oxidative stress consequently induced tier-2 and tier-3 oxidative stress responses. Based on our observations made at chronic exposure period (90 d) after the single i.v. injection of ceria ENM, we could extend the model of oxidative stress hierarchy mechanisms for ceria-ENM-induced toxicity. Considering the evaluation of all the oxidative stress indices measured in 3-brain regions, oxidative stress effects were more prominent in hippocampus and the least in cerebellum, but no specific pattern or any significant difference was deduced.

Adaptive designs can improve the efficiency of drug development, but further research is needed before some are more widely implemented. One such design is a treatment-selection design, which begins with k treatment arms, but only a subset is carried forward after an interim analysis. The final analysis of the selected arm(s) is then performed using the data from both stages of the study. One issue with this design is ensuring the Type I error rate is controlled, but there have been a number of proposals that largely address this. A second drawback that has not yet been fully addressed is that the maximum likelihood estimate of the selected arm at the final analysis is often biased upward due to the selection method. Unbiased estimators already exist for this design, but methods with an acceptable balance between bias and mean squared error (MSE) are lacking. In this dissertation, two estimation approaches are proposed. The first is a parametric bootstrap resampling method in which the level of bias adjustment applied is driven by a comparison of the observed results to those expected when all arms have equal true means. The second approach is an empirical Bayes estimator that implements a novel limited translation function. These methods are compared to previously proposed approaches with respect to bias and MSE for studies that have either a normal or binomial endpoint. Both proposed methods are shown to exhibit reduced bias with reasonable MSE in some simulated scenarios, but the resampling method consistently shows similar, or improved, performance compared to previous approaches across the examined scenarios. The utility of this resampling method is further demonstrated by showing that it can be implemented when the arm with the second largest mean is selected for stage 2. It is also shown that the resampling method can be extended to when more than one arm is selected in stage 1, when there is a futility analysis, or when the study has a time-to-event endpoint. Recommendations on confidence intervals are also provided. The results demonstrate that the parametric bootstrap resampling method is a viable estimation approach for treatment-selection designs.

Les cancers du sein et de la prostate sont, de par leur fréquence, au 1er rang de l’ensembledes cancers chez la femme et chez l’homme. Au stade métastatique, la maladie devient incurable etles traitements possèdent uniquement une action palliative. Ainsi, de nombreuses études tentent dedécouvrir de nouveaux traitements et de nouvelles associations thérapeutiques afin d’améliorer et derallonger la survie de ces patients. Parmi ces approches, la curcumine, composé polyphénoliqueextrait du curcuma, est un agent préventif largement étudié in vitro et in vivo mais peu exploré enclinique curative. De plus, l’apport de thérapies personnalisées grâce à l’approfondissement descaractéristiques biologiques de la tumeur aide également à cibler des traitements de plus en plusefficaces.L’objectif de ce travail a été d’optimiser la prise en charge thérapeutique du cancer du seinet de la prostate métastatiques. Ainsi, nous avons simultanément développé une base de données etdes études cliniques de phase II.Deux études rétrospectives ont été menées dans le cancer du sein métastatique à partir d’unebase de données de 756 patientes afin d’étudier des aspects peu connus en pratique clinique. Lapremière (n=511) étudiait le rôle des paramètres tumoraux initiaux sur la survie métastatique. Enanalyse multivariée, l’âge, le grade histologique selon Scarff-Bloom Richardson et le nombre deganglions envahis au diagnostic initial gardent leur valeur pronostique après la rechute métastatiqueavec une survie métastatique plus longue pour les patientes jeunes, les tumeurs de bas grade ou sansenvahissement ganglionnaire.La deuxième étude (n=529) examinait l’intérêt de traiter les patientes au-delà de 3 lignes dechimiothérapie. Cette analyse a montré un bénéfice en terme de survie de poursuivre lachimiothérapie en lignes tardives, pour les patientes aptes à la recevoir.Deux essais prospectifs ont étudié la combinaison du docétaxel et de la curcumine. Ils fontsuite à une étude de phase I réalisée au Centre Jean Perrin. Dans le premier essai de phase II, nonrandomisé, 30 patients atteints d’un cancer de la prostate métastatique résistant à la castration ontbénéficié de cette association en 1ière ligne de traitement. L’association docétaxel et curcumine apermis d’obtenir une réponse partielle pour 40% des cas et une réponse du PSA pour 59% despatients. Les gélules de curcumine (6 g/j) ont été bien tolérées et les patients ont présenté une trèsbonne compliance. De plus, cette association semble être active à la fois chez les patients ayant oun’ayant pas un profil d’agressivité neuroendocrine. L’analyse des marqueurs neuroendocriniens aégalement retenu la chromogranine A comme le plus pertinent, en complément du PSA, pour lesuivi des patients. La deuxième étude prospective, de phase II randomisée, comparait le docétaxelen monothérapie à la combinaison (docetaxel/curcumine), en 1ière ou 2ème ligne de traitement ducancer du sein métastatique. Cette étude est actuellement en cours avec 21 patientes incluses sur les100 initialement prévues. En parallèle, une analyse de biomarqueurs sériques, cibles de lacurcumine (CRP, ICAM et VEGF), a été réalisée. Le nombre de patientes incluses est actuellementtrop faible pour pouvoir constater un bénéfice de l’ajout de la curcumine au docétaxel.Ces travaux soulignent l’importance des lignes de chimiothérapies et des paramètrestumoraux initiaux dans la prise en charge du cancer du sein métastatique. Ils apportent égalementdes résultats prometteurs quant à l’utilisation de la curcumine pour potentialiser le docétaxel dans lecancer de la prostate résistant à la castration et dans le cancer du sein métastatique. / Breast and prostate cancer are the most common cancer in women and men. Metastatic diseaseremains incurable and therapies are only palliative. Thus, many studies try to discover newtreatments and new combination therapies to improve and extend the survival of these patients.Among these strategies, curcumin, a dietary polyphenol derived from curcuma, was found topossess cancer preventive properties in vitro and in vivo, but has been less studied clinically.Moreover, personalized therapies, by intensifying the knowledge of biological characteristics of thetumor, may also help to target more effective therapies.The objective of this work was to optimize the therapeutic management of breast and prostatecancers. Hence, we have developed simultaneously a database and phase II clinical trials.A database of 756 patients with metastatic breast cancer were achieved two retrospective studies.The first (n=511) examined the impact of initial tumor parameters on metastatic survival.Multivariate analysis retained age, Scarff-Bloom Richardson grade and axillary lymph nodeinvolvement as significant independent prognostic factors after relapse with a longer metastaticsurvival for young patients, low grade tumor or no axillary lymph node involvement.The second study (n=529) investigated the interest to treat patients with more than 3 chemotherapylines. This analysis has shown a benefit of each line in terms of OS, for patients able to receive it.Two prospective clinical trials have studied the association docetaxel plus curcumin, following aCentre Jean Perrin phase I study. In the first phase II trial, non-randomized, 30 patients, withcastration-resistant prostate cancer, have been treated with the combination in first treatment line.Objective response rate was obtained for 40% and PSA response for 59% of patients. Curcumin (6g/d) has been well tolerated and patients had higher curcumin capsules compliance. Furthermore,this therapeutic combination seemed to be active in patients with or without the aggressiveneuroendocrine characteristics. In parallel, chromogranin A has been selected as the bestneuroendocrine marker to follow patient, in addition to PSA. The second phase II trial, randomizedcompare docetaxel in monotherapy to the association docetaxel plus curcumin, in first and secondtreatment ligne in metastatic breast cancer. This study is ongoing with 21 patients enrolled on the100 originally planned. In parallel, we have realized an analysis of curcumin target solublebiomarkers (CRP, ICAM and VEGF). However, the number of patients included is presently notlarge enough to evidence any benefit of adding curcumin to docetaxel.These data support chemotherapy lines interest on survival and the impact of initial tumorparameters on metastatic breast cancer therapeutic management. Moreover, the first results ofcurcumin and docetaxel encourage further studies with this therapeutic association.

On constate que la majorité des essais de phase III, conduits après des essais de phase II pourtant prometteurs, sont "négatifs", la nouvelle thérapeutique se révélant finalement trop toxique ou insuffisamment efficace. L'hétérogénéité de la population participant aux différentes phases de développement est une explication. Elle induirait une estimation erronée de la toxicité et, par dilution de l'effet traitement, conduirait à arrêter l'évaluation thérapeutique alors que peut être un sous-ensemble de cette population, définie à partir d'une caractéristique particulière, pourrait en bénéficier.Dans cette thèse, nous proposons dans un premier temps une réflexion sur les aspects méthodologiques des essais de phase II qui permettraient d'améliorer l'identification précoce des thérapeutiques toxiques et des populations les plus sensibles et donc de ne planifier des essais de phase III que sur des populations encore mieux ciblées. Dans un second temps, nous présentons une nouvelle méthodologie d'essai de phase II que nous avons développée pour prendre en compte l'hétérogénéité de la population et son intérêt en pratique clinique courante. Avec cette méthode, qui est une extension du plan de Fleming à deux étapes, le développement des médicaments est moins fréquemment arrêté pour la population entière et moins de patients non sensibles à la nouvelle thérapeutique sont exposés à des molécules potentiellement toxiques, durant l'étape 2 de l'essai de phase II ou plus tard lors de l'essai de phase III.

Acrylamide is an industrially produced chemical with known neurotoxic, reproductive toxin and carcinogenic effects. The carcinogenicity associated with acrylamide is mostly attributed to its metabolism by liver CYP2E1. However, studies investigating the effects of acrylamide on CYP2E1 enzyme are limited. In this study, it was aimed to investigate in vivo interaction of carcinogenic acrylamide on microsomal cytochrome P450 enzyme activities, and protein levels, and on cytosolic NQO1 and GST enzyme activities of rabbit liver, kidney and lung of acrylamide-treated rabbits. The in vivo protective effect of resveratrol, a phenolic compound, was also investigated on acrylamide toxicity. New Zealand male rabbits were treated with acrylamide and resveratrol, separately in different doses and conditions. Their combined effects were also investigated. CYP2E1-dependent p-Nitrophenol hydroxylase, NDMA N-demethylase and aniline 4-hydroxylase activities were found to be significantly increased in acrylamide-treated rabbit liver (1.80-3.0 fold) and kidney (1.6-fold). Rabbit liver and kidney CYP2E1 protein levels (determined by western blot analyisis) also increased approximately 2-fold due to acrylamide treatment. In rabbit liver, resveratrol was found significantly effective in decreasing both acrylamide-induced CYP2E1-dependent enzyme activities (approximately 1.5-1.80 fold) and CYP2E1 protein levels (approximately 1.5-1.70 fold). Additionally, resveratrol significantly decreased acrylamide-induced CYP2E1 protein level (2-2.5 fold) in rabbit kidney. However, no significant change was observed in rabbit lung CYP2E1-dependent enzyme activities and CYP2E1 protein levels due to acrylamide, resveratrol or their combined treatments. Furthermore, it was found that acrylamide treatment significantly increased CYP3A6-dependent erythromycin N-demethylase enzyme activity (1.85-fold) and CYP3A6 protein levels in rabbit liver (1.69-fold). No change was observed in CYP2B4-dependent benzphetamine N-demethylase enzyme activities of rabbit liver, kidney and lung by in vivo acrylamide, resveratrol or their combined treatments. Moreover, total GST and GST-Mu activities of rabbit kidney (1.5-fold, respectively) and total GST activity of rabbit lung (1.6-fold) were increased significantly only in resveratrol treated group. NQO1 enzyme activity of rabbit kidney was significantly increased by acrylamide treatment (1.6-fold). The results of the present study have demonstrated for the first time that acrylamide induces rabbit liver and kidney CYP2E1-dependent enzyme activities and CYP2E1 protein levels. The induction of CYP2E1 enzyme activity and protein level by acrylamide treatment can stimulate formation of other toxic compounds and procarcinogens metabolized by CYP2E1 which in turn further potentiates the risk of hepatotoxicity, mutagenicity and carcinogenicity. In the present study, it was also demonstrated for the first time that acrylamide treatment also increases CYP3A6 enzyme activity in rabbit liver which may lead to alterations in drug metabolism. The results of this study have also suggested that resveratrol may have protective effects on acrylamide induced toxicity / however, further in vivo studies are required to clarify the effect of resveratrol on both acrylamide-induced toxicity and anti-oxidant enzymes.

Phase II clinical trials aim to potentially screen out ineffective and identify effective therapies to move forward to randomized phase III trials. Single-arm studies remain the most utilized design in phase II oncology trials, especially in scenarios where a randomized design is simply not practical. Due to concerns regarding excessive toxicity or ineffective new treatment strategies, interim analyses are typically incorporated in the trial, and the choice of statistical methods mainly depends on the type of primary endpoints. For oncology trials, the most common primary objectives in phase II trials include tumor response rate (binary endpoint) and progression disease-free survival (time-to-event endpoint). Interim strategies are well-developed for both endpoints in single-arm phase II trials. The advent of molecular targeted therapies, often with lower toxicity profiles from traditional cytotoxic treatments, has shifted the drug development paradigm into establishing evidence of biological activity, target modulation and pharmacodynamics effects of these therapies in early phase trials. As such, these trials need to address simultaneous evaluation of safety as well as proof-of-concept of biological marker activity or changes in continuous tumor size instead of binary response rates. In this dissertation, we extend a predictive probability design for binary outcomes in the single-arm clinical trial setting and develop two interim designs for continuous endpoints, such as continuous tumor shrinkage or change in a biomarker over time. The two-stage design mainly focuses on the futility stopping strategies, while it also has the capacity of early stopping for efficacy. Both optimal and minimax designs are presented for this two-stage design. The multi-stage design has the flexibility of stopping the trial early either due to futility or efficacy. Due to the intense computation and searching strategy we adopt, only the minimax design is presented for this multi-stage design. The multi-stage design allows for up to 40 interim looks with continuous monitoring possible for large and moderate effect sizes, requiring an overall sample size less than 40. The stopping boundaries for both designs are based on predictive probability with normal likelihood and its conjugated prior distributions, while the design itself satisfies the pre-specified type I and type II error rate constraints. From simulation results, when compared with binary endpoints, both designs well preserve statistical properties across different effect sizes with reduced sample size. We also develop an R package, PPSC, and detail it in chapter four, so that both designs can be freely accessible for use in future phase II clinical trials with the collaborative efforts of biostatisticians. Clinical investigators and biostatisticians have the flexibility to specify the parameters from the hypothesis testing framework, searching ranges of the boundaries for predictive probabilities, the number of interim looks involved and if the continuous monitoring is preferred and so on.

BackgroundAtrazine suppression of the LH surge slowly develops over time and peaks after 4 days; sensitivity to atrazine decreases after 8 or 14 days of dosing. Adaptation of the LH response was correlated with increased phase I and phase II liver enzyme activity/expression. MethodsThe effect of atrazine on the LH surge was evaluated in female Sprague-Dawley rats administered 100 mg/kg/day atrazine by gavage for 1, 2, 3, or 4 consecutive days or 6.5, 50, or 100 mg/kg/day atrazine for 4, 8, or 14 days. ResultsNo statistically significant effects of atrazine were seen on peak plasma LH or LH area under the curve (AUC) after one, two, or three doses of 100 mg/kg/day. Four daily doses of 50 or 100 mg/kg atrazine significantly reduced peak LH and LH AUCs, whereas 6.5 mg/kg/day had no effect. After 8 or 14 days of treatment, statistically significantly reduced peak LH and LH AUC were observed in the 100 mg/kg/day dose group, but not in the 6.5 or 50 mg/kg/day dose groups, although significantly reduced LH was observed in one sample 9 hr after lights-on in the 50 mg/kg/day dose group on day 14. The number of days of treatment required to achieve a significant suppression of the LH surge is consistent with the repeat-dose pharmacokinetics of the chlorotriazines. ConclusionThe apparent adaptation to the effect of atrazine on the LH surge after 8 or 14 days may be related to the induction of phase I or, more likely, phase II metabolism observed in this study after 8 days, or to a decreased sensitivity of the hypothalamic-pituitary-adrenal axis or an homeostatic adaption of the effect of atrazine on the LH surge mechanism. Birth Defects Research 110:246-258, 2018. (c) 2017 The Authors. Birth Defects Research Published by Wiley Periodicals, Inc.