Receptor tyrosine kinases (RTK) are key regulators of growth, differentiation and survival of epithelial cells and play a significant role in the development and progression of cancers derived from these tissues. In malignant cells, these receptors and their downstream signalling pathways are often deregulated, leading to cell hyper-proliferation, enhanced cell survival and increased metastatic potential. Furthermore, endothelial expressed RTKs regulate tumor angiogenesis allowing for tumor growth and maintenance by promoting their vascularization. Epithelial malignancies such as squamous cell carcinomas (SCC), non-small cell lung (NSCLC) and malignant mesotheliomas have very limited treatment options when presenting as metastatic disease. RTKs, particularly the epidermal growth factor (EGFR) and the vascular endothelial growth factor (VEGFR) receptors, have been shown to play significant roles in the pathogenesis of these tumor types. Statins are potent inhibitors of HMG-CoA reductase, the rate limiting enzyme of the mevalonate pathway, that are widely used as hypercholesterolemia treatments. The mevalonate pathway produces a variety of end products that are critical for many different cellular pathways, thus, targeting this pathway can affect multiple signalling pathways. Our laboratory has previously shown that lovastatin can induce tumor specific apoptosis especially in SCC and that 23% of recurrent SCC patients treated with lovastatin as a single agent showed disease stabilization in our Phase I clinical trial. Subsequently, our lab was able to demonstrate that lovastatin in combination with gefitinib, a potent inhibitor of the EGFR showed co-operative cytotoxicity when combined (Chapter 2). Furthermore, the pro-apoptotic and cytotoxic effects of these agents were found to be synergistic and to be manifested in several types of tumor cell lines including SCC, NSCLC and glioblastoma. I was able to expand upon these important findings and demonstrated that lovastatin, through its ability to disrupt the actin cytoskeleton, inhibited EGFR dimerization and activation (Chapter 3). This novel mechanism targeting this receptor has clinical implications as lovastatin treatment combined with gefitinib showed co-operative inhibitory effects on EGFR activation and downstream signalling. The RTK family of proteins share similar features with respect to activation, internalization and downstream signalling effectors. I further demonstrated that lovastatin can inhibit the VEGFR-2 in endothelial cells and mesotheliomas, where VEGF and its receptor are co-expressed driving their proliferation, and induces synergistic cytotoxicity in mesothelioma cells in combination with VEGFR-2 tyrosine kinase inhibitors (Chapter 4). These findings suggest that statins may augment the effects of a variety of RTK inhibitors in a similar fashion representing a novel combinational therapeutic approach in a wide repertoire of human cancers. More importantly, based on this work, we initiated a Phase I/II study evaluating high dose rosuvastatin and the EGFR inhibitor tarceva in SCC and NSCLC patients at our institute. This clinical evaluation will provide invaluable data that will play a role in developing this novel therapeutic strategy. Together, the work embodied in this thesis provides a model for the regulation of EGFR/VEGFR-2 activation and signalling by targeting the rho family of proteins that demonstrates a novel mechanism that can be exploited to refine current therapeutic paradigms.

Receptor tyrosine kinases (RTK) are key regulators of growth, differentiation and survival of epithelial cells and play a significant role in the development and progression of cancers derived from these tissues. In malignant cells, these receptors and their downstream signalling pathways are often deregulated, leading to cell hyper-proliferation, enhanced cell survival and increased metastatic potential. Furthermore, endothelial expressed RTKs regulate tumor angiogenesis allowing for tumor growth and maintenance by promoting their vascularization. Epithelial malignancies such as squamous cell carcinomas (SCC), non-small cell lung (NSCLC) and malignant mesotheliomas have very limited treatment options when presenting as metastatic disease. RTKs, particularly the epidermal growth factor (EGFR) and the vascular endothelial growth factor (VEGFR) receptors, have been shown to play significant roles in the pathogenesis of these tumor types. Statins are potent inhibitors of HMG-CoA reductase, the rate limiting enzyme of the mevalonate pathway, that are widely used as hypercholesterolemia treatments. The mevalonate pathway produces a variety of end products that are critical for many different cellular pathways, thus, targeting this pathway can affect multiple signalling pathways. Our laboratory has previously shown that lovastatin can induce tumor specific apoptosis especially in SCC and that 23% of recurrent SCC patients treated with lovastatin as a single agent showed disease stabilization in our Phase I clinical trial. Subsequently, our lab was able to demonstrate that lovastatin in combination with gefitinib, a potent inhibitor of the EGFR showed co-operative cytotoxicity when combined (Chapter 2). Furthermore, the pro-apoptotic and cytotoxic effects of these agents were found to be synergistic and to be manifested in several types of tumor cell lines including SCC, NSCLC and glioblastoma. I was able to expand upon these important findings and demonstrated that lovastatin, through its ability to disrupt the actin cytoskeleton, inhibited EGFR dimerization and activation (Chapter 3). This novel mechanism targeting this receptor has clinical implications as lovastatin treatment combined with gefitinib showed co-operative inhibitory effects on EGFR activation and downstream signalling. The RTK family of proteins share similar features with respect to activation, internalization and downstream signalling effectors. I further demonstrated that lovastatin can inhibit the VEGFR-2 in endothelial cells and mesotheliomas, where VEGF and its receptor are co-expressed driving their proliferation, and induces synergistic cytotoxicity in mesothelioma cells in combination with VEGFR-2 tyrosine kinase inhibitors (Chapter 4). These findings suggest that statins may augment the effects of a variety of RTK inhibitors in a similar fashion representing a novel combinational therapeutic approach in a wide repertoire of human cancers. More importantly, based on this work, we initiated a Phase I/II study evaluating high dose rosuvastatin and the EGFR inhibitor tarceva in SCC and NSCLC patients at our institute. This clinical evaluation will provide invaluable data that will play a role in developing this novel therapeutic strategy. Together, the work embodied in this thesis provides a model for the regulation of EGFR/VEGFR-2 activation and signalling by targeting the rho family of proteins that demonstrates a novel mechanism that can be exploited to refine current therapeutic paradigms.

Hintergrund und Fragestellung: Seit der Zulassung des Tyrosinkinase-Inhibitors (TKI) Imatinib im Jahre 2001 hat sich die Therapie der chronisch myeloischen Leukämie (CML) grundlegend verändert. Imatinib inhibiert die konstitutiv aktive Tyrosinkinase BCR-ABL, welche die verstärkte Proliferation der leukämischen Zellen und die Entwicklung der CML bedingt. Das sehr gute klinische Ansprechen auf eine Imatinib-Behandlung resultierte in einer beschleunigten Zulassung dieses TKI auch bei pädiatrischen Patienten im Jahre 2003. Aufgrund von Punktmutationen und/oder strukturellen Änderungen innerhalb des BCR-ABL Fusionsproteins können sich Resistenzen gegenüber Imatinib entwickeln. Deshalb wurden Zweit- und Drittgenerations TKI wie Dasatinib und Bosutinib entwickelt. Imatinib wirkt nicht hoch spezifisch und hemmt neben BCR-ABL auch weitere Tyrosinkinasen, wie z.B. c-KIT, PDGF-R und c-FMS, welche am Knochenstoffwechsel beteiligt sind. Die Stimulation des Rezeptors c-FMS bewirkt die Differenzierung monozytärer Vorläuferzellen in knochenabbauende Osteoklasten. Zusätzlich unterliegt die Entwicklung der knochenaufbauenden Osteoblasten spezifischen Signalkaskaden an denen PDGF-R und c-Abl beteiligt sind. Als Nebenwirkung einer TKI-Therapie beeinträchtigt die Inhibition dieser Signaltransduktionswege somit das Knochen-“Remodelling“, indem die Entwicklung und funktionelle Aktivität von Osteoklasten reduziert wird. Gleichzeitig wird die Aktivität von Osteoblasten gestärkt, aber deren Proliferation inhibiert. Diese Dysbalance von Knochenaufbau und -abbau mit gestörter Kalziumhomöostase bedingt bei erwachsenen CML-Patienten veränderte endokrinologische Parameter des Kalziumhaushaltes, eine vermehrte Knochenmineralisation und eine erhöhte trabekuläre Knochendichte. Dagegen wurden bei pädiatrischen CML-Patienten unter Imatinib-Therapie Längenwachstumsstörungen beobachtet, welche bezüglich des Wirkmechanismuses von Imatinib auf den wachsenden Knochen bis heute noch nicht im Detail geklärt sind. Spekulativ ist auch, ob Zweit- und Drittgenerations-TKI ebenso wie Imatinib den Knochenstoffwechsel bei pädiatrischen Patienten stören. Angelehnt an ein erfolgreiches Applikationsschema bei erwachsenen CML-Patienten steht zusätzlich die Frage im Raum, ob eine intermittierende Gabe von TKIs (einen Monat Therapie, einen Monat Pause) eine Minderung der Nebenwirkung auf den Knochen bewirken könnte, ohne die Wirkung auf die CML-Behandlung zu beeinträchtigen. Vor diesem Hintergrund wurde ein Nagermodel etabliert, um Nebenwirkungen auf den Knochenstoffwechsel unter TKI-Exposition zu analysieren. Junge, wachsende Ratten wurden hierzu vom präpubertären Alter bis zur Adoleszenz kontinuierlich oder intermittierend mit den TKIs Imatinib, Dasatinib und Bosutinib exponiert und die Wirkung auf das wachsende Skelettsystem untersucht. Methoden: 4 Wochen alte männliche Wistar Ratten wurden über einen Zeitraum von 10 Wochen chronisch mit jeweils einem der drei im Trinkwasser gelösten TKIs exponiert. Neben einer unbehandelten Kontrollkohorte erhielt eine Gruppe jeweils eine Standarddosis und eine hohe Dosis (entsprechend der doppelten Standarddosis) des entsprechenden TKIs kontinuierlich. Eine weitere Gruppe erhielt die hohe Dosis intermittierend (an drei aufeinanderfolgenden Tagen TKI, danach vier Tage nur Wasser). Die Konzentrationen im Trinkwasser betrugen für Imatinib 1 mM und 2 mM und für Dasatinib und Bosutinib jeweils 50 µM und 100 µM. Nach zweiwöchiger (präpubertär), vierwöchiger (pubertäres Stadium) und zehnwöchiger Exposition (postpubertär) wurden die Tiere aller Gruppen nekropsiert und Röhrenknochen, Lendenwirbel und Blut asserviert. Zur Beurteilung des Knochenmetabolismus wurden folgende Parameter erhoben: Knochenlängen, Knochendichten mittels pQCT, trabekuläre Strukturen mittels µCT, Knochenfestigkeit mittels des 3-Punkt-Biege-Test und endokrinologische Parameter im Serum mittels ELISA. Zusätzlich wurde der jeweilige TKI Serum-Spiegel bestimmt. Ergebnisse: Die Gewichtsentwicklung, körperliche Entwicklung und das Sozialverhalten zeigten keine Unterschiede beim Vergleich von Kontrollkohorten mit exponierten Tieren. Die kontinuierliche Exposition mit Imatinib und Dasatinib bewirkte dosisabhängig eine Reduktion der Knochenlängen der Femura und der Tibiae. Bosutinib zeigte diesen Effekt nicht. Die intermittierende Exposition mit hoher Dosis resultierte in einer Knochenlängenreduktion, welche exakt dem Effekt der Standarddosis entsprach. Weiterhin resultierte aus der Exposition mit Imatinib oder Dasatinib eine Verminderung der trabekulären Knochendichten der Femura und Tibiae im präpubertären Stadium. Ratten, welche hoch dosiert Imatinib erhielten, zeigten diese Reduktion ebenfalls im pubertären Stadium, nicht jedoch unter Dasatinib- und Bosutinib-Exposition. Postpubertär unterschieden sich die trabekulären Dichten von Femura und Tibiae der exponierten Gruppen nicht von den Kontrollkohorten. Auf die kortikale Knochendichte und die kortikale Dicke dieser Röhrenknochen zeigte sich kein messbarer Effekt der TKI. Dennoch trat - nur nach Exposition der hohen Imatinibdosis - eine signifikant verminderte femorale Bruchfestigkeit postpubertär auf. Am Lendenwirbelkörper war pubertär und postpubertär die Höhe unter Imatinib-Exposition vermindert, während die Gesamt- und kortikale Knochendichte präpubertär erhöht war bei tendenziell erniedrigter trabekulärer Knochendichte. Die kortikale Dicke wurde durch alle TKI nicht beeinflusst. Dasatinib und Bosutinib bewirkten keinen Effekt auf die Wirbelhöhe, aber eine tendenzielle Minderung der trabekulären Knochendichte. Der serologisch erfassbare Knochenresorptionsmarker „tatrate resistant acidic phosphatase“ (TRAP) war unter kontinuierlicher Exposition mit hoher Dosis von Imatinib zu allen Zeitpunkten erniedrigt. Postpubertär zeigte sich dieser Effekt auch unter Standard- und Hochdosis von Bosutinib. Der Knochenformationsmarker Osteocalcin war unter Imatinib bei allen Kohorten zu allen Analysezeitpunkten erniedrigt, während Dasatinib und Bosutinib keinen Effekt auf diesen Parameter zeigten. Die erfassten Serum-Hormonparameter (Wachstumshormon, Parathormon) lagen unter der Exposition mit Imatinib als erhöhte Wachstumshormonspiegel pubertär und als verminderte Parathormonspiegel prä- und pubertär vor. Unter der Exposition mit Dasatinib kam es ebenfalls pubertär zu einer Erhöhung der Wachstumshormonspiegel und präpubertär zu einer tendenziellen Erhöhung der Parathormonspiegel. Postpubertär normalisierten sich beide Parameter unter der Exposition mit Imatinib und Dasatinib wieder. Unter Bosutinib konnte nur postpubertär erniedrigte Parathormonspiegel ermittelt werden. Eine intermittierende TKI-Exposition resultierte in einem Aufholwachstum und einer teilweise Normalisierung der knochenspezifischen Serumparameter. Als wichtige unerwartete Nebenwirkung zeigte sich unter Langzeitexposition mit Imatinib und Dasatinib eine Zunahme des Herzgewichtes. Unter Imatinib resultierten daraus keine klinischen Auffälligkeiten, während unter Dasatinib eine Herzinsuffizienz zum Tod eines Tieres führte. Bosutinib zeigte keine weiteren makropathologisch erfassbaren Nebenwirkungen. Bis heute sind keine kardialen Nebenwirkungen bei pädiatrischen Patienten nach mehrjähriger TKI-Therapie publiziert. Schlussfolgerung: Das etablierte juvenile Nagertiermodell ist gut geeignet, um die Nebenwirkungen einer Langzeitexposition von TKI auf den wachsenden Knochen zu erfassen. Bei Kindern und Adoleszenten klinisch beschriebene Wachstumsretardierungen unter Imatinib ließen sich zweifelsfrei bei Ratten verifizieren. Bei fehlenden klinischen Daten von Kindern zu Dasatinib präjudiziert das Modell, dass Dasatinib so wie Imatinib den gleichen, Bosutinib hingegen kaum einen Effekt auf den Knochen ausübt. Eine intermittierende Gabe der TKI scheint die Nebenwirkungen auf den Knochen abzumildern und könnte eine neue Möglichkeit der TKI-Therapie für pädiatrische Patienten darstellen. Aus dem Tiermodell der Langzeit-exponierten juvenilen Ratte lässt sich ableiten, dass beim wachsenden Kind unter jahrelanger TKI-Therapie klinisch sorgfältig der Knochenstoffwechsel und das Längenwachstum überwacht und unter Dasatinib zusätzlich kardiale Nebenwirkungen beachten werden sollten. / Background: Since its approval in 2001 the tyrosine kinase inhibitor (TKI) imatinib has revolutionized the therapy of chronic myeloid leukaemia (CML). Imatinib inhibits the constitutively active tyrosine kinase (TK) BCR-ABL causing the increased proliferation of the leukemic cells and the progress of CML. According to improved survival rates imatinib has been licensed as frontline therapy also for paediatric CML in 2003. However, due to point mutations or structural changes within the BCR-ABL fusion protein resistance to imatinib occurs. Therefore 2nd and 3rd generation TKI like dasatinib and bosutinib have been developed. Beside BCR-ABL, Imatinib exerts also off-target effects on further TKs like c-KIT, PDGF-R, c-FMS which are involved in bone metabolism. Stimulation of the receptor c-FMS leads to the differentiation of monocytic progenitors to bone resorbing osteoclasts. In addition, the development of bone forming osteoblasts underlies specific signalling cascades involving PDGF-R and c-Abl. As a side effect of TKI therapy these specific signalling cascades are inhibited impairing bone remodelling by reducing the development and functional activity of osteoclasts. Simultaneously osteoblasts’ differentiation is promoted while their proliferation is inhibited. This dysbalance of bone formation and resorption results in altered endocrinological serum markers of the calcium homeostasis, increased bone mineralization, and increased trabecular bone density in adult CML patients. In contrast paediatric CML patients show longitudinal growth retardations under imatinib therapy, however, the detailed action of imatinib on the growing bone is not clarified yet. Additionally, it is unclear if 2nd and 3rd generation TKI will also disturb bone metabolism in paediatric CML patients. Based on an effective treatment strategy in adult CML patients, it is also questioned if intermittent TKI treatment (one month “on”, one month “off”) could minimise side effects on the bone without impairing CML therapy. On this background a rodent model was established to study side effects of TKI treatment on bone metabolism. Juvenile growing rats where exposed from prepubertal age till adolescence continuously or intermittently to imatinib, dasatinib, and bosutinib and the effects on the growing skeleton were analysed. Methods: Four weeks old male Wistar rats were chronically exposed to varying concentrations of one of the three TKIs via the drinking water for 10 weeks. Besides untreated controls a standard dosage group and a high dosage group (equalling the twofold standard dose) received every TKI continuously, while an additional group received the high dosage TKI in an intermittent fashion (3 days per week: “on” TKI; 4 days water without TKI). The concentrations applied were 1 mM and 2 mM for imatinib and 50 µM and 100 µM each for dasatinib and bosutinib, respectively. After 2 weeks (prepubertal), 4 weeks (pubertal stage), and 10 weeks (postpubertal) of exposure, respectively, animals were sacrificed and long bones, lumbar vertebra and blood were isolated. To evaluate bone metabolism the following parameters were analysed: bone length, bone mineral density (BMD) by pQCT, trabecular structure by µCT, bone strength by 3-point bending test, and endocrinological parameters by ELISA. Additionally, serum levels of TKIs were investigated. Results: In comparison to controls no alterations of exposed animals’ bodyweight, overall development and social behaviour were observed. Continuous exposure of imatinib and dasatinib led dose dependently to reduced femoral and tibial length. No such effect was observed under bosutinib. Intermitted exposure of high-dose TKIs resulted in reduced effects on femoral and tibial length identical to the effect observed in groups receiving just standard dose. Furthermore, exposure of imatinib and dasatinib lowered femoral and tibial trabecular BMD prepubertally. Rats receiving high dose imatinib showed reduced femoral and tibial trabecular BMD at pubertal stage, while this effect was not observed under dasatinib and bosutinib exposure. Postpubertally, femoral and tibial trabecular BMD of all exposed groups did not differ from controls. Femoral and tibial cortical BMD and cortical thickness were not affected by TKI exposure. However, under high dose imatinib exposure femoral mechanical breaking strength was reduced postpubertally. In vertebra the height was reduced under imatinib exposure pubertally and postpubertally, while the total and cortical BMD were increased prepubertally and trabecular BMD tended to be reduced. Cortical thickness was not affected by any TKI tested. Dasatinib and bosutinib exhibited no effect on the height of the vertebra but trabecular BMD tended to be reduced. The serum bone resorption marker ‘tartrate resistant acidic phosphatase’ (TRAP) was found reduced under continuous exposure of high dose of imatinib at all time points tested. Postpubertally, the same effect was detected after standard and high dosage of bosutinib. The bone formation marker osteocalcin was reduced in all groups and at all time points tested under imatinib exposure, whereas no such effect was observed for dasatinib and bosutinib. Serum bone related hormone markers (growth hormone (GH) and parathyroid hormone (PTH)) revealed under imatinib exposure increased GH levels pubertally whereas PTH was reduced pre- und pubertally. During dasatinib exposure GH levels were elevated pubertally and PTH levels were increased prepubertally. Postpubertally, both parameters normalised again under imatinib and dasatinib exposure. During bosutinib exposure reduced PTH levels were detected postpubertally only. Intermitted TKI exposure resulted in catch-up growth and partial normalisation of bone specific serum parameters. As major unexpected side effect during exposure increasing heart weights could be observed under long-time imatinib and dasatinib exposure. No clinical changes were observed under imatinib, whereas dasatinib led to cardiac insufficiency leading to death of one animal. Bosutinib showed no additional macrospathologic assessable side effects. To date no cardiac side effects were published in paediatric patients under prolonged TKI therapy. Conclusion: The established juvenile rat model is appropriate to examine side effects of long-term TKI exposure on the growing bone. Published longitudinal growth retardation in children and adolescents under imatinib treatment could be unequivocally mimicked in this rat model. Due to not yet available clinical experience with dasatinib in paediatric patients, this model predicts that dasatinib alters bone metabolism like imatinib whereas bosutinib shows less detectable effects. Intermitted TKI treatment may reduce side effects on the growing bone and therefore could represent a new opportunity of TKI therapy for paediatric patients. Summing up, TKI long-term exposure in this juvenile rat model challenges physicians to diligently monitor bone metabolism in not outgrown paediatric patients during long-term TKI treatment and additionally assess cardiac side effects under dasatinib exposure.

Diss. (sammanfattning) Umeå : Umeå universitet, 2009. / Härtill 4 uppsatser.

As we age, our ability to cope with a variety of stresses significantly decreases. One of the features of an ageing organism is the dramatic increase in the number of cells arrested in the G1 phase, a process known as senescence. It is well established that the senescence phenotype leads to a change in the way cells respond to stress. However, the molecular mechanisms by which these cells cope and/or respond to a variety of environmental challenges remain unknown. In general, cells respond to stress by engaging a variety of mechanisms; one of them is the assembly of cytoplasmic foci known as stress granules (SGs). These entities are considered as part of the survival pathways that are activated at the beginning of any stress to protect key cellular elements which allow a quick recovery if the stress is rapidly removed. However, we do not know whether SGs formation is activated during senescence. In this study, we investigated the formation and the role of SGs in senescent cells exposed to various stresses. We demonstrated that while SGs can assemble in response to oxidative stress (OS) during all the steps leading to senescence activation, their number significantly increases at late stage of senescence. This increase correlates with a rapid decrease in the expression of the cyclin kinase inhibitor p21, one of the main players in the activation of the senescence phenotype. Although the OS-induced recruitment of p21 mRNA to SGs correlates with a significant increase in its half-life, this translocation interferes with p21 translation only at late senescence. This translation inhibition could be explained by the co-recruitment of CUGBP1, a known translation activator during senescence of p21, and p21 mRNA to SGs. Therefore, our data suggest that SGs formation and the reduction in p21 protein levels represent two main events through which senescent cells respond to stress conditions.

Receptor tyrosine kinases (RTK) are key regulators of growth, differentiation and survival of epithelial cells and play a significant role in the development and progression of cancers derived from these tissues. In malignant cells, these receptors and their downstream signalling pathways are often deregulated, leading to cell hyper-proliferation, enhanced cell survival and increased metastatic potential. Furthermore, endothelial expressed RTKs regulate tumor angiogenesis allowing for tumor growth and maintenance by promoting their vascularization. Epithelial malignancies such as squamous cell carcinomas (SCC), non-small cell lung (NSCLC) and malignant mesotheliomas have very limited treatment options when presenting as metastatic disease. RTKs, particularly the epidermal growth factor (EGFR) and the vascular endothelial growth factor (VEGFR) receptors, have been shown to play significant roles in the pathogenesis of these tumor types. Statins are potent inhibitors of HMG-CoA reductase, the rate limiting enzyme of the mevalonate pathway, that are widely used as hypercholesterolemia treatments. The mevalonate pathway produces a variety of end products that are critical for many different cellular pathways, thus, targeting this pathway can affect multiple signalling pathways. Our laboratory has previously shown that lovastatin can induce tumor specific apoptosis especially in SCC and that 23% of recurrent SCC patients treated with lovastatin as a single agent showed disease stabilization in our Phase I clinical trial. Subsequently, our lab was able to demonstrate that lovastatin in combination with gefitinib, a potent inhibitor of the EGFR showed co-operative cytotoxicity when combined (Chapter 2). Furthermore, the pro-apoptotic and cytotoxic effects of these agents were found to be synergistic and to be manifested in several types of tumor cell lines including SCC, NSCLC and glioblastoma. I was able to expand upon these important findings and demonstrated that lovastatin, through its ability to disrupt the actin cytoskeleton, inhibited EGFR dimerization and activation (Chapter 3). This novel mechanism targeting this receptor has clinical implications as lovastatin treatment combined with gefitinib showed co-operative inhibitory effects on EGFR activation and downstream signalling. The RTK family of proteins share similar features with respect to activation, internalization and downstream signalling effectors. I further demonstrated that lovastatin can inhibit the VEGFR-2 in endothelial cells and mesotheliomas, where VEGF and its receptor are co-expressed driving their proliferation, and induces synergistic cytotoxicity in mesothelioma cells in combination with VEGFR-2 tyrosine kinase inhibitors (Chapter 4). These findings suggest that statins may augment the effects of a variety of RTK inhibitors in a similar fashion representing a novel combinational therapeutic approach in a wide repertoire of human cancers. More importantly, based on this work, we initiated a Phase I/II study evaluating high dose rosuvastatin and the EGFR inhibitor tarceva in SCC and NSCLC patients at our institute. This clinical evaluation will provide invaluable data that will play a role in developing this novel therapeutic strategy. Together, the work embodied in this thesis provides a model for the regulation of EGFR/VEGFR-2 activation and signalling by targeting the rho family of proteins that demonstrates a novel mechanism that can be exploited to refine current therapeutic paradigms.

Le cancer du sein était au premier rang des cancers chez la femme en termes d'incidence et de mortalité en 2008 (GLOBOCAN 2012, IARC). A l'origine de cette mortalité on observe notamment le développement de résistances aux chimiothérapies conventionnelles mais aussi le développement des métastases qui sont responsables de 90% des décès des patientes. Il est donc indispensable d'étendre l'arsenal thérapeutique à la disposition des cliniciens avec de nouvelles molécules actives sur les formes de cancers résistants aux chimiothérapies et ciblant le processus métastatique.Au sein de notre équipe nous avons identifié un nouvel inhibiteur de la LIM Kinase (LIMK) baptisé « Pyr1 ». La LIMK est responsable de la régulation de la dynamique des microtubules et des microfilaments d'actine. Dans un premier article (Prudent et al., Cancer Research, 2012) nous avons montré que Pyr1 avait une activité anti-mitotique et anti-migratoire dans un modèle cellulaire de cancer de l'utérus. De plus une analyse pilote a montré que Pyr1 était efficace dans un modèle murin de leucémie où il augmente la survie des souris. Étant donné que la LIMK est surexprimée dans les cancers du sein et qu'une des conséquences cellulaires d'un traitement par Pyr1 est une stabilisation des microtubules, comme le fait le paclitaxel couramment utilisé pour le traitement des formes invasives de ces cancers , l'objectif de ma thèse a été d'étudier l'efficacité thérapeutique de Pyr1 sur des modèles cellulaires et murins de cancer du sein, et notamment sur les cancers résistants au paclitaxel.Nous avons montré que Pyr1 ralentissait la croissance des tumeurs primaires et réduisait leur taille. Pyr1 est bien toléré et l'effet anti-tumoral est également observé sur des modèles résistants au paclitaxel. Nous avons ensuite étudié l'effet de Pyr1 sur la migration et l'invasion des cellules tumorales en utilisant notamment la microscopie intra-vitale. Nous avons observé que, bien qu'il ne diminue pas la vitesse de migration des cellules tumorales, Pyr1 entraine in vivo un changement morphologique important. Enfin, Pyr1 n'empêche pas la dispersion métastatique mais il prévient de manière efficace la croissance des métastases.Nous en avons conclu que Pyr1 est une molécule intéressante pour le traitement des cancers du sein résistants aux chimiothérapies conventionnelles et également sur le traitement des métastases. / Breast cancer is the most common diagnosed cancer in women worldwide with an increase of 20% and 14% in terms of incidence and mortality, respectively, in 2008 (GLOBOCAN, 2012). This increase is mainly due to the lack of therapeutics that target the development of metastasis responsible for 90% of cancer death. Moreover, the development of resistance to available chemotherapies limits their effectiveness. It's an urgent need to find new drugs that target the metastatic process and are efficient on resistant cancers.Our team has identified a new LIM Kinase (LIMK) inhibitor called “Pyr1”. LIM Kinases are implicated in the dynamic regulation of the actin and microtubule cytoskeleton. We previously published data showing that Pyr1 has an anti-mitotic and anti-migratory activity on a cervical cancer cell line. Moreover, a pilot in vivo study has shown that Pyr1 was efficient in a leukemia mouse model where it increases the lifespan of treated compared to control mice (Prudent et al., 2012).LIM Kinases have been shown to be overexpressed in breast cancer. Moreover the chemotherapeutical agents currently used for this kind of cancer belong to the class of taxanes (such as paclitaxel). Taxanes are cytotoxic compound that directly binds to microtubules and stabilizes them. Since Pyr1 treatment also results in microtubules stabilization, with a complete different mechanism of action, we have decided to investigate the anti-cancer effect of Pyr1 on breast cancer cell lines and xenografts, including paclitaxel resistant models.We showed that Pyr1 decreases primary tumor growth and reduces their size. Pyr1 is well tolerated and the anti-tumor effect is also observed on paclitaxel resistant models. We then studied Pyr1 effects on migration and invasion in vitro and in vivo. Intravital imaging of tumors showed that, whereas Pyr1 didn't slow down tumor cell migration, it induced a cell morphological change. Finally, Pyr1 does not affect metastasis spreading but prevents their growth.These results indicate that LIMK inhibitors, such as Pyr1, may represent a pharmacological alternative for taxanes resistant tumors. Moreover, they could be potent agents to reduce the size of metastasis.

Nous étudions une nouvelles familles d'inhibiteurs de kinase: les benzopyridoindole. Ces molécules ont des effets antiprolifératifs sur des lignées cancéreuses et représentent les têtes de série de possibles agents anti-cancéreux. We study on a new family of kinase inhibitors: benzopyridoindole. These molecules have antiproliferative effects on cancer cell lines and represent the lead of potential anti-cancer products. / Benzo[e]pyridoindoles are novel potent inhibitors of aurora kinases. We performed a SAR study to improve their activity and water solubility. Amino-benzo[e]pyridoindolones were found to be potent hydrosoluble anti-proliferative molecules. They induced a massive arrest in mitosis, prevented histone H3 phosphorylation as well as disorganizing the mitotic spindles. Upon a delay, cells underwent binucleated and finally died. Taking into account their interesting preclinal characteristics, their efficiency towards xenografts in nude mice and their apparent safety in animals, these molecules are promising new anti-cancer drugs. They probably target a metabolic signaling pathway, besides aurora B inhibition. In addition to their possible applications, these inhibitors are tools for cell biology studies. C4, a low ATP affinity inhibitor of aurora B kinase, revealed that the basal activity of the kinase is required for histone H3 phosphorylation in prophase and for chromosome compaction in anaphase. These waves of activation/deactivation of the kinase, during mitosis, corresponded to different conformations of the passenger chromosomal complex.

Receptor tyrosine kinases (RTK) are key regulators of growth, differentiation and survival of epithelial cells and play a significant role in the development and progression of cancers derived from these tissues. In malignant cells, these receptors and their downstream signalling pathways are often deregulated, leading to cell hyper-proliferation, enhanced cell survival and increased metastatic potential. Furthermore, endothelial expressed RTKs regulate tumor angiogenesis allowing for tumor growth and maintenance by promoting their vascularization. Epithelial malignancies such as squamous cell carcinomas (SCC), non-small cell lung (NSCLC) and malignant mesotheliomas have very limited treatment options when presenting as metastatic disease. RTKs, particularly the epidermal growth factor (EGFR) and the vascular endothelial growth factor (VEGFR) receptors, have been shown to play significant roles in the pathogenesis of these tumor types. Statins are potent inhibitors of HMG-CoA reductase, the rate limiting enzyme of the mevalonate pathway, that are widely used as hypercholesterolemia treatments. The mevalonate pathway produces a variety of end products that are critical for many different cellular pathways, thus, targeting this pathway can affect multiple signalling pathways. Our laboratory has previously shown that lovastatin can induce tumor specific apoptosis especially in SCC and that 23% of recurrent SCC patients treated with lovastatin as a single agent showed disease stabilization in our Phase I clinical trial. Subsequently, our lab was able to demonstrate that lovastatin in combination with gefitinib, a potent inhibitor of the EGFR showed co-operative cytotoxicity when combined (Chapter 2). Furthermore, the pro-apoptotic and cytotoxic effects of these agents were found to be synergistic and to be manifested in several types of tumor cell lines including SCC, NSCLC and glioblastoma. I was able to expand upon these important findings and demonstrated that lovastatin, through its ability to disrupt the actin cytoskeleton, inhibited EGFR dimerization and activation (Chapter 3). This novel mechanism targeting this receptor has clinical implications as lovastatin treatment combined with gefitinib showed co-operative inhibitory effects on EGFR activation and downstream signalling. The RTK family of proteins share similar features with respect to activation, internalization and downstream signalling effectors. I further demonstrated that lovastatin can inhibit the VEGFR-2 in endothelial cells and mesotheliomas, where VEGF and its receptor are co-expressed driving their proliferation, and induces synergistic cytotoxicity in mesothelioma cells in combination with VEGFR-2 tyrosine kinase inhibitors (Chapter 4). These findings suggest that statins may augment the effects of a variety of RTK inhibitors in a similar fashion representing a novel combinational therapeutic approach in a wide repertoire of human cancers. More importantly, based on this work, we initiated a Phase I/II study evaluating high dose rosuvastatin and the EGFR inhibitor tarceva in SCC and NSCLC patients at our institute. This clinical evaluation will provide invaluable data that will play a role in developing this novel therapeutic strategy. Together, the work embodied in this thesis provides a model for the regulation of EGFR/VEGFR-2 activation and signalling by targeting the rho family of proteins that demonstrates a novel mechanism that can be exploited to refine current therapeutic paradigms.

Philosophiae Doctor - PhD / Globally, breast cancer is the most common cancer affecting women and it is predicted that in 2030 about 12 million deaths will occur with approximately 21.7 million new cases [2]. Genetic risk factors as well as race and ethnicity, account for about 5-10% of all breast cancer occurrences. Triple negative breast cancer (TNBC), tumors that tested negative for oestrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor 2 (HER2), contribute to 10-20% of all breast carcinomas [3,4] and is known to be a more aggressive type of cancer with varying degree of response to chemotherapeutic and radiation therapy [5,6] / 2022-02-24