@phdthesis{Hoffmann2017,
  author    = {Hoffmann, Helene},
  title     = {Identifying regulators of tumor vascular morphology},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-142348},
  school      = {Universit{\"a}t W{\"u}rzburg},
  year      = {2017},
  abstract  = {In contrast to normal vessels, tumor vasculature is structurally and functionally abnormal. Tumor vessels are highly disorganized, tortuous and dilated, with uneven diameter and excessive branching. Consequently, tumor blood flow is chaotic, which leads to hypoxic and acidic regions in tumors. These conditions lower the therapeutic effectiveness and select for cancer cells that are more malignant and metastatic. The therapeutic outcome could be improved by increasing the functionality and density of the tumor vasculature. Tumor angiogenesis also shows parallels to epithelial to mesenchymal transition (EMT), a process enabling metastasis. Metastasis is a multi-step process, during which tumor cells have to invade the surrounding host tissue to reach the circulation and to be transported to distant sites. We hypothesize that the variability in the phenotype of the tumor vasculature is controlled by the differential expression of key transcription factors. Inhibiting these transcription factors might be a promising way for angiogenic intervention and vascular re-engineering. Therefore, we investigated the interdependence of tumor-, stroma- and immune cell-derived angiogenic factors, transcription factors and resulting vessel phenotypes. Additionally, we evaluated whether transcription factors that regulate EMT are promising targets for vascular remodeling. We used formalin fixed paraffin embedded samples from breast cancer patients, classified according to estrogen-, progesterone- and human epidermal growth factor receptor (HER) 2 status. Establishing various techniques (CD34 staining, laser microdissection, RNA isolation and expression profiling) we systematically analyzed tumor and stroma-derived growths factors. In addition, vascular parameters such as microvessel size, area, circularity and density were assessed. Finally the established expression profiles were correlated with the observed vessel phenotype. As the SNAI1 transcriptional repressor is a key regulator of EMT, we examined the effect of vascular knockdown of Snai1 in murine cancer models (E0771, B16-F10 and lewis lung carcinoma). Among individual mammary carcinomas, but not among subtypes, strong differences of vascular parameters were observed. Also, little difference between lobular carcinomas and ductal carcinomas was found. Vessel phenotype of Her2 enriched carcinomas was similar to that of lobular carcinomas. Vessel morphology of luminal A and B and basal-like tumors resembled each other. Expression of angiogenic factors was variable across subtypes. We discovered an inverse correlation of PDGF-B and VEGF-A with vessel area in luminal A tumors. In these tumors expression of IL12A, an inhibitor of angiogenesis, was also correlated with vessel size. Treatment of endothelial cells with growth factors revealed an increased expression of transcription factors involved in the regulation of EMT. Knockdown of Snai1 in endothelial cells of mice increased tumor growth and decreased hypoxia in the E0771 and the B16-F10 models. In the lewis lung carcinomas, tumor vascularity and biodistribution of doxorubicin were improved. Here, doxorubicin treatment in combination with the endothelial cell-specific knockdown did slow tumor growth. This shows that SNAI1 is important for a tumor's vascularization, with the significance of its role depending on the tumor model. The methods established in this work open the way for the analysis of the expression of key transcription factors in vessels of formalin fixed paraffin embedded tumors. This research enables us to find novel targets for vascular intervention and to eventually design novel targeted drugs to inhibit these targets.},
  subject      = {Antiangiogenese},
  language  = {en}
}
@phdthesis{Kaymak2019,
  author    = {Kaymak, Irem},
  title     = {Identification of metabolic liabilities in 3D models of cancer},
  doi       = {10.25972/OPUS-18154},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-181544},
  school      = {Universit{\"a}t W{\"u}rzburg},
  year      = {2019},
  abstract  = {Inefficient vascularisation of solid tumours leads to the formation of oxygen and nutrient gradients. In order to mimic this specific feature of the tumour microenvironment, a multicellular tumour spheroid (SPH) culture system was used. These experiments were implemented in p53 isogenic colon cancer cell lines (HCT116 p53 +/+ and HCT116 p53-/-) since Tp53 has important regulatory functions in tumour metabolism. First, the characteristics of the cells cultured as monolayers and as spheroids were investigated by using RNA sequencing and metabolomics to compare gene expression and metabolic features of cells grown in different conditions. This analysis showed that certain features of gene expression found in tumours are also present in spheroids but not in monolayer cultures, including reduced proliferation and induction of hypoxia related genes. Moreover, comparison between the different genotypes revealed that the expression of genes involved in cholesterol homeostasis is induced in p53 deficient cells compared to p53 wild type cells and this difference was only detected in spheroids and tumour samples but not in monolayer cultures. In addition, it was established that loss of p53 leads to the induction of enzymes of the mevalonate pathway via activation of the transcription factor SREBP2, resulting in a metabolic rewiring that supports the generation of ubiquinone (coenzyme Q10). An adequate supply of ubiquinone was essential to support mitochondrial electron transport and pyrimidine biosynthesis in p53 deficient cancer cells under conditions of metabolic stress. Moreover, inhibition of the mevalonate pathway using statins selectively induced oxidative stress and apoptosis in p53 deficient colon cancer cells exposed to oxygen and nutrient deprivation. This was caused by ubiquinone being required for electron transfer by dihydroorotate dehydrogenase, an essential enzyme of the pyrimidine nucleotide biosynthesis pathway. Supplementation with exogenous nucleosides relieved the demand for electron transfer and restored viability of p53 deficient cancer cells under metabolic stress. Moreover, the mevalonate pathway was also essential for the synthesis of ubiquinone for nucleotide biosynthesis to support growth of intestinal tumour organoids. Together, these findings highlight the importance of the mevalonate pathway in cancer cells and provide molecular evidence for an enhanced sensitivity towards the inhibition of mitochondrial electron transfer in tumour-like metabolic environments.},
  subject      = {Tumor},
  language  = {en}
}
@phdthesis{Marquardt2023,
  author    = {Marquardt, Andr{\´e}},
  title     = {Machine-Learning-Based Identification of Tumor Entities, Tumor Subgroups, and Therapy Options},
  doi       = {10.25972/OPUS-32954},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-329548},
  school      = {Universit{\"a}t W{\"u}rzburg},
  year      = {2023},
  abstract  = {Molecular genetic analyses, such as mutation analyses, are becoming increasingly important in the tumor field, especially in the context of therapy stratification. The identification of the underlying tumor entity is crucial, but can sometimes be difficult, for example in the case of metastases or the so-called Cancer of Unknown Primary (CUP) syndrome. In recent years, methylome and transcriptome utilizing machine learning (ML) approaches have been developed to enable fast and reliable tumor and tumor subtype identification. However, so far only methylome analysis have become widely used in routine diagnostics. The present work addresses the utility of publicly available RNA-sequencing data to determine the underlying tumor entity, possible subgroups, and potential therapy options. Identification of these by ML - in particular random forest (RF) models - was the first task. The results with test accuracies of up to 99\% provided new, previously unknown insights into the trained models and the corresponding entity prediction. Reducing the input data to the top 100 mRNA transcripts resulted in a minimal loss of prediction quality and could potentially enable application in clinical or real-world settings. By introducing the ratios of these top 100 genes to each other as a new database for RF models, a novel method was developed enabling the use of trained RF models on data from other sources. Further analysis of the transcriptomic differences of metastatic samples by visual clustering showed that there were no differences specific for the site of metastasis. Similarly, no distinct clusters were detectable when investigating primary tumors and metastases of cutaneous skin melanoma (SKCM). Subsequently, more than half of the validation datasets had a prediction accuracy of at least 80\%, with many datasets even achieving a prediction accuracy of - or close to - 100\%. To investigate the applicability of the used methods for subgroup identification, the TCGA-KIPAN dataset, consisting of the three major kidney cancer subgroups, was used. The results revealed a new, previously unknown subgroup consisting of all histopathological groups with clinically relevant characteristics, such as significantly different survival. Based on significant differences in gene expression, potential therapeutic options of the identified subgroup could be proposed. Concludingly, in exploring the potential applicability of RNA-sequencing data as a basis for therapy prediction, it was shown that this type of data is suitable to predict entities as well as subgroups with high accuracy. Clinical relevance was also demonstrated for a novel subgroup in renal cell carcinoma. The reduction of the number of genes required for entity prediction to 100 genes, enables panel sequencing and thus demonstrates potential applicability in a real-life setting.},
  subject      = {Maschinelles Lernen},
  language  = {en}
}
@phdthesis{Roehrig2015,
  author    = {R{\"o}hrig, Florian},
  title     = {Verbesserung der Medikamenteneinbringung in solide Tumoren durch Modifikation der extrazellul{\"a}ren Matrix},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-117381},
  school      = {Universit{\"a}t W{\"u}rzburg},
  year      = {2015},
  abstract  = {Bei der Behandlung solider Tumoren spielen systemisch verabreichte Chemotherapeutika eine wich- tige Rolle. Allerdings akkumulieren diese Therapeutika besser in normalem Gewebe als in Tumoren. Als Ursache f{\"u}r diesen unzureichenden Transport von Medikamenten in den Tumor wurde bisher vor allem die dysfunktionale Tumorvaskulatur diskutiert. Diese befindet sich in einem chaotischen und unreifen Zustand ohne ausreichende Bedeckung der Gef{\"a}ße mit stabilisierenden Perizyten. Aus dem Zustand der Vaskulatur resultierend erreichen Medikamente den Tumor nur in geringem Ausmaß und werden dort heterogen verteilt. Als Grund f{\"u}r den Zustand der Vaskulatur wur- de ein großer {\"U}berschuss an pro-angiogenetischen Faktoren im Tumor ausgemacht. Durch eine anti-angiogenetische Behandlung konnte in pr{\"a}klinischen Modellen f{\"u}r einen gewissen Zeitraum die Tumorvaskulatur „normalisiert" werden. Dies zeichnete sich vor allem durch Ver{\"a}nderung von zwei wichtigen Parametern f{\"u}r die Medikamenteneinbringung aus: zum Einen kommt es zu einer Reduktion der Gef{\"a}ßdichte. Zum Anderen zu einer Reifung der Blutgef{\"a}ße. In einem Teil von Pati- enten scheint dabei der Effekt der Gef{\"a}ßverbesserung zu {\"u}berwiegen und es kann eine verbesserte Perfusion detektiert werden. Mutmaßlich f{\"u}hrt dies auch zu einer verbesserten Einbringung von Therapeutika in den Tumor und so zu einer erh{\"o}hten Effizienz der Therapie. In einem weiteren Teil der Patienten scheint jedoch der Effekt der Gef{\"a}ßreduktion zu {\"u}berwiegen und die detektierte Perfusion im Tumor wird durch die Behandlung verringert. Das in dieser Arbeit verwendete MT6-Fibrosarkom-Modell reagierte auf eine anti-angiogenetische Therapie nicht mit einer sonst in murinen Modellen beobachteten Wachstumsreduktion. Die- se erm{\"o}glichte eine so bisher nicht m{\"o}gliche Untersuchung der sekund{\"a}ren Effekte einer anti- angiogenetischen Therapie wie die Medikamenteneinbringung in den Tumor. Die Vaskulatur in MT6-Tumoren zeigte dabei nach einer anti-angiogenetischen Vorbehandlung, die erwarteten Merk-male einer „normalisierten" Vaskulatur wie eine Reduktion der Gef{\"a}ßdichte bei gleichzeitiger Rei- fung der verbleibenden Gef{\"a}ße. Dies f{\"u}hrte jedoch nicht zu einer verbesserten Effizienz einer subsequenten Chemotherapie. Durch Vergleich mit einem weiteren Tumor-Modell, dem 4T1-Modell f{\"u}r ein metastasierendes Mammakarzinom, konnten signifikante Unterschiede im Gef{\"a}ßbild beider Modelle ausgeschlossen werden. Durch mikroskopische Methoden konnte dabei beobachtet werden, dass die Diffusion von Medikamenten aus den Blutgef{\"a}ßen des MT6-Modells im Vergleich zum 4T1-Modell verringert war. Weitere Untersuchungen deuten auf eine Differenz in der Qualit{\"a}t der extrazellul{\"a}ren Matrix der verwendeten Tumor-Modelle. Durch mRNA-Expressionsanalysen konnte die Enzymfamilie der Lysyloxidasen als m{\"o}gliche Ursache f{\"u}r diesen Diffusionsunterschied identi- fiziert werden. Lysyloxidasen katalysieren vor allem die Quervernetzung von Proteinen der Extra- zellul{\"a}rmatrix. Im Weiteren konnte gezeigt werden, dass die Quervernetzung von Matrixproteinen durch Lysyloxidasen urs{\"a}chlich f{\"u}r die Diffusions-Inhibierung kleiner Molek{\"u}le wie das Chemo- therapeutikum Doxorubicin sein kann. Durch spezifische Inhibition der Lysyloxidasen mittels des Inhibitors βAPN konnte diese Diffusions-Inhibition sowohl in vitro als auch im MT6-Tumor-Modell nahezu vollst{\"a}ndig verhindert werden. Die hohe Aktivit{\"a}t von Lysyloxidasen im MT6-Modell stell- te allerdings kein Alleinstellungsmerkmal dieses Modells dar. In weiteren Untersuchungen konnte gezeigt werden, dass Lysyloxidasen in einer Vielzahl von murinen und humanen Tumorzelllinien {\"u}berexprimiert wird. Die Inhibition von Lysyloxidasen durch βAPN konnte dabei in allen unter- suchten Modellen die Einbringung von Medikamenten in den Tumor erh{\"o}hen und k{\"o}nnte so eine sinnvolle adjuvante Maßnahme zur Verbesserung bestehender Chemotherapien darstellen.},
  subject      = {Lysin-Oxidase},
  language  = {de}
}