<p> One of the greatest problems in prostate cancer management today is accurate identification of patients who require treatment for aggressive disease versus those with indolent disease who are suitable for observational strategies. Histological appearance of the tumor, called Gleason score in the prostate cancer field, is the most predictive measure currently used. However, recent studies in multiple tumor types have shown that histological appearance does not always reflect the underlying molecular phenotype of the lesion. Therefore, in prostate cancer specifically, assessment of a molecular marker of androgen receptor driven epithelial differentiation may have clinical predicative capabilities. SAM pointed domain-containing Ets transcription factor (SPDEF) is a potential AR target gene that has shown to be necessary and sufficient for epithelial cell differentiation in many tissues. Although generally associated with good prognosis, SPDEF's role in cancer in unclear. This study demonstrates, through retrospective immunohistochemical analysis, the utility of SPDEF as a predictive biomarker for patients that have an extended benefit from androgen deprivation therapy (ADT). Furthermore, dual roles of SPDEF to inhibit the initiation and supporting the progression of castrate recurrent disease through novel androgen receptor expression regulation in castrate conditions are shown. In ADT na&iuml;ve patients, SPDEF did not associate with metastatic disease or an induction of epithelial to mesenchymal transition. However, aggressive tumors tended to be larger, have greater SPDEF variability, and lack vimentin expression; a phenotype that could be explained by a partial EMT. In conclusion, SPDEF may be clinically useful to assess the epithelial phenotype of tumors, and could have utility identifying patients that will respond well to androgen deprivation therapy.</p>

Tumor cell arrest and the formation of stable adhesive interactions between tumor cells and endothelial cells or underlying matrix in the microvasculature are crucial steps in the metastatic process. A sensitive hydrodynamic adhesion assay was used to investigate the regulation of melanoma cell adhesion stabilization to the extracellular matrix protein fibronectin, and also to study the relationship between tumor cell adhesion stabilization and organ-specific blood borne metastasis. Modulators of intracellular signals were used to determine the role of the signaling pathways in regulation of adhesion stabilization. Modulators of intracellular (Ca$\sp{2+}$) were found to inhibit stabilization, but extracellular Ca$\sp{2+}$ was not required. Inhibitors of calmodulin, protein kinase C, tyrosine kinases, and the actin cytoskeleton also inhibited adhesion stabilization. Phorbol esters, cAMP modulators, and G protein inhibitors had no effect on adhesion stabilization. Most of the drug treatments that inhibited adhesion stabilization also had significant effects on the actin cytoskeleton organization of the melanoma cells. These results suggest a role for intracellular (Ca$\sp{2+}$), calmodulin, protein kinase C, and tyrosine kinases in the regulation of melanoma cell adhesion stabilization. Adhesion stabilization of metastatic murine RAW117 large-cell lymphoma cells to the extracellular matrix proteins fibronectin and vitronectin, several RGD containing peptides, and microvascular endothelial cells from the liver or lung was also investigated. The highly liver metastatic RAW117-H10 subline showed faster adhesion stabilization to fibronectin, vitronectin, and RGD peptides than did the poorly metastatic RAW117-P cells or the lung- and liver-metastatic RAW117-L17 cells. The adhesion stabilization of the RAW117-H10 cells to the extracellular matrix proteins and RGD peptides was inhibited by anti-$\beta\sb3$ integrin monoclonal antibodies and RGD peptides. In contrast, the RAW117-L17 subline had the shortest stabilization time to unstimulated microvascular endothelial cells of the lung and hepatic sinusoids, followed by RAW117-H10 cells and RAW117-P cells. RGD peptides and monoclonal antibodies against the Mac-1 and $\beta\sb3$ integrins did not inhibit adhesion stabilization of RAW117-H10 cells to endothelial cells. These results suggest that different metastatic variants of large-cell lymphoma cells use different mechanisms to adhere in the microcirculation.

The compelling evidence that platelets play a contributory role in hematogenous metastasis raises a need to elucidate the molecular mechanisms by which tumor cells activate and adhere to platelets. To model the in vivo interactions of platelets with arrested tumor cells, heparinized mepacrine-labeled whole blood was perfused over an A875 melanoma cell monolayer in a parallel plate flow chamber at shear rates characteristic of the microvasculature. Epi-fluorescence video microscopy utilizing a cooled CCD camera and digital image processing were used to quantify the dynamic adhesion and aggregation of platelets. Varying the exposure time of the monolayer to the 445 nm excitation wavelength revealed that photoactivation was stimulating the tumor cells and that the cells were relatively nonthrombogenic in absence of the light over the time scale of the flow experiments. Preactivation of the tumor cells was attempted with 12(S)-HETE, thrombin, and TNF-alpha and resulted in close to no platelet deposition.

Tumor cell migration and invasion of body tissues are prerequisite mediators for lymphatic or hematogenous cancer dissemination. To date, there is insufficient understanding of what triggers the metastatic cascade, and of how the interplay among cell receptors, the cellular and acellular components of the extracellular matrix and proteolytic enzymes mediate cancer migration, invasion, proliferation and survival. In addition to the inherent complexity of each one of the aforementioned phenomena is the lack of an experimental technique capable of dissecting the mechanisms that mediate the dynamic invasive and migratory behavior at the cellular level and with respect to the properties of the cell environment. The goal of my thesis was to develop an automated system for cell tracking in three dimensions and use it to model the dynamics of cancer invasion and migration. Therefore the hardware and software were designed for a fully automated optical 3D cell tracking system that quantified long-term invasion and migration of cancer cells infiltrating 3D extracellular matrix analogs. The quantitative analysis of the cell trajectories employed a novel formulation of the continuous Markov model that evaluated the potential for invasive or lateral motion and cell stops. The infiltration of human HT1080 fibrosarcoma and human MDA-MB-231 adenocarcinoma cells, was monitored in plain or Matrigel-containing collagen type I gels. Parameters such as the speed subpopulations, the persistence of motion in certain directions, the turning frequency of the cells, the preferred directions of motion, and the invasion depth profiles over time quantified infiltration at the cellular level. Distinct migratory and invasive phenotypes significantly dependent on the gel composition were identified for the two cell types. The HT1080 cell line expressed a high motility phenotype and well-preserved lateral motion on the plain collagen gel surface. The basement membrane components transformed the HT1080 cells to robust invaders by significantly enhancing the matrix infiltration and the turning frequency. The low motility, slow invasion and low turning behavior of MDA-MB-231 cells indicated that their invasiveness may depend on matrix-degrading activity. To the best of my knowledge this is the first study employing a detailed set of quantitative descriptors to demonstrate that tumor invasion and migration are dynamic processes of individual cells that depend significantly on the cell type and the tumor microenvironment.

This dissertation addresses certain statistical issues in two biomedical fields, namely, modeling breast cancer screening program and correlated survival data analysis. For the breast cancer screening project, this study investigates statistical approaches to quantitatively describing the age effect on screening sensitivity and sojourn time distribution. Such an investigation is directly motivated by the need to understand the inherent relationships between age and these important quantities. Age effect is incorporated through generalized linear models under a progressive disease modeling framework. Parameter estimates are obtained by maximizing conditional likelihood functions. Among a set of potential models, the Akeike's information criterion and likelihood ratio test are used in model selection and inferences. Extensive simulation studies show that the estimators have reasonable accuracy and the model selection criterion works well. The proposed methods are illustrated using data from two large breast cancer screening trials. For correlated survival data analysis, an interesting yet often ignored problem is considered, that is when cluster sizes may be informative to the outcome of interest, based on a within-cluster resampling approach and a weighted marginal model. Large sample properties for the within-cluster resampling estimators are derived under the Cox proportional hazards model, including the consistency and asymptotic normality of the regression coefficient estimators and the weak convergence property of the estimated baseline cumulative hazard function. The weighted marginal model is constructed by incorporating the inverse of cluster sizes as weights in the estimating equations. Simulation studies are conducted to assess and compare the finite-sample behaviors of the estimators and the proposed methods are applied to a dental data example as an illustration.

One of the elements commonly seen in cancer is the change in methylation status of the genome. These aberrations in methylation appear to be critical for the neoplastic phenotype and manifest as changes to gene expression of oncogenes and tumour suppressors. In addition to epigenetic alterations, the proteins involved in maintaining the plastic methylation status of the genome, DNA methyltransferases and demethylases, also show methylation-independent protein-protein interactions that have effects on cell cycle progression and proliferation. As changes in gene expression and mitotic regulation are seminal elements of cancer, and because several methylated DNA binding proteins show differential expression in a wide variety of cancers, these proteins serve as prime targets for anticancer therapies. This thesis relates to exploring both current and forthcoming possibilities and mechanisms of utilizing the DNA methylation machinery for pharmacological intervention of cancer. Chapter two deals with an antisense drug, currently in clinical trials, targeted to reduction of DNA methyltransferase 1, the maintenance methylation enzyme in mammalian cells. Our data indicate that the existence of a common truncation mutation of the adenomatous polyposis coli gene seen in some forms of sporadic and familial colorectal cancer may lead to downstream upregulation of DNA methyltransferase 1, as reconstitution of the wildtype protein reduces DNA methyltransferase 1 mRNA and protein. Reduction of the transcripts of this methylation enzyme with an antisense oligonucleotide decreases the tumourigenicity of these colorectal cancer cells, and provides a rationale for use of this drug in colorectal cancer patients and prophylactic treatment of adenomatous polyposis coli mutation-bearing individuals. Chapter three describes the rationale, design, and in vitro and in vivo testing of antisense molecules against the methylated DNA binding protein MBD2. These drugs red

1,25-dihydroxyvitaminD$ sb3$ (1,25(OH)$ sb2$D$ sb3$) is anti-proliferative and pro-differentiative in a variety of cell types, including human keratinocytes. Exposure of cultured normal human keratinocytes to 1,25(OH)$ sb2$D$ sb3$ markedly reduces ($ sp3$H) thymidine incorporation and cell number and arrests these cells in the G1/G0 phase of the cell cycle. Calcium and 1,25(OH)$ sb2$D$ sb3$ act in concert to modulate the expression of two important cell-cycle associated genes, c-fos and p53, and of markers of keratinocyte differentiation. In contrast, inhibition of cell growth and of the cell cycle associated oncogene, c-myc, in the malignant keratinocyte cell line, HPK1Aras, requires 10 to 100 fold higher concentrations of 1,25(OH)$ sb2$D$ sb3$ than do the immortal non-malignant human keratinocyte cell line, HPK1A. Cell cycle analysis also reveals that 10-100 fold higher concentrations of 1,25(OH)$ sb2$D$ sb3$ are required to induce cell cycle arrest in HPK1Aras cells as compared to HPK1A cells. Analysis of the vitamin D receptor (VDR) from these two cell lines reveals identical sizes, numbers, and ligand binding characteristics. Furthermore, the sequence of the DNA binding domain of this receptor is unchanged in the vitamin D resistant HPK1Aras cells. Gel mobility shift assays using extracts from both cell lines reveals that the complexes formed by HPK1Aras nuclear extracts in the presence of a vitamin D DNA response element (VDRE) probe contain VDR but not its dimerization partner, the retinoic acid X receptor (RXR). In contrast, HPK1A nuclear extracts form complexes that contain both VDR and RXR. Overexpressing wild type RXR$ alpha$ in HPK1Aras cells results in VDRE-binding complexes containing VDR/RXR heterodimers. However, the sequence of the RXR$ alpha$ is unchanged in HPK1Aras cells and its expression in both cell lines is the same. Western blot analysis of RXR$ alpha$ from ras transformed keratinocytes suggests that its post-translational modification is

The studies included in this thesis were aimed at identifying the mechanisms that lead to aberrant expression of the PTHRP gene in cancer. / We have used the hepatocyte growth factor receptor oncogene, Tpr-Met, as a model and examined the effect of this oncogene on PTHRP expression. When transfected into Fisher rat 313 (Fr3T3) fibroblasts, Tpr-Met increased the transcription of PTHRP mRNA and secretion of the protein. To identify the signaling pathways involved we analyzed a mutant of Tpr-Met, Tyr489, that was impaired in activating a number of downstream effectors, including Phosphatidylinositol-3 kinase, Grb2 and Shc. The ability of Tpr-Met/Tyr 489 mutant to induce PTHRP expression was significantly reduced. Furthermore, inhibiting Ras using lovastatin, in wild-type Tpr-Met transfected cells, Completely Suppressed PTHRP levels, suggesting that the mechanism was Ras-dependent. / We next directly investigated the effect of Ras on PTHRP expression in vitro, and on hypercalcemia of malignancy in vivo. When transfected PTHRP cells the activated mutant of Ras (RasV12) potently increased PTHRP mRNA and protein levels. When RasV12 expressing cells were subcutaneously injected into BALB/c/nu/ nu mice, the tumors developed rapidly, and signs of hypercalcemia were detected within 2 weeks. Inhibiting Ras using a specific inhibitor, B-1096, completely blocked expression of PTHRP, in vitro, and suppressed the sips of hypercalcemia in vivo. These results show that inhibiting Ras was sufficient to block tumor expression of PTHRP and development of hypercalcemia. / Using rat Leydig tumor H-500 cells, we next investigated effector pathways downstream of Ras that mediate serum stimulated PTHRP expression. PTHRP mRNA was decreased by a dominant negative mutant of Raf (Raf C4B) and by a MEK inhibitor (PD 098059), implicating the involvement of Ras-Raf-MEK pathway in the serum response. In addition, stimulation with UV light or expression of an activated form of Rac (Rac V12) was sufficient to increase PTHRP mRNA. Furthermore, a dominant negative mutant Of Rac (Rac N17) also blocked serum induced expression of mRNA. This suggests that the stress-activated pathways may provide alternative mechanisms that can regulate the PTHRP gene. These pathways also appear to be important in the serum induced response. Collectively, the results from these studies contribute to our limited knowledge of the mechanisms governing PTHRP expression in cancer. The findings also provide novel targets to explore for improved therapy of hypercalcemia.

The human progranulin gene encodes a 593 amino acid secreted glycoprotein, progranulin (also called PC cell-derived growth factor, acrogranin or granulin/epithelin precursor). The biological activities of this protein were poorly defined at the beginning of the project. Here, we determined that the progranulin gene is widely expressed in vivo and strongly associated with immune, neuronal and highly proliferative epithelial cells. It is not expressed in fibroblasts and endothelial cells in vivo unless these cells are stimulated by, for example, tissue damage and this expression correlates temporally with the healing process. Thus progranuhn is induclbly expressed in mesenchymal cells but is a constitutive product in proliferative epithelial cells. These studies defined a physiological context for investigating the roles of progranulin in the body. To correlate the expression studies with function, we assessed the ability of progranulin to regulate wound repair and epithelial growth. Progranulin enhances the growth and motility of fibroblasts and endothelial cells, and promotes the formation of tubule-like structures by endothelial cells on Matrigel, suggesting that progranulin may accelerate fibroplasia and serve as a pro-angiogenic factor in the proper environment. These effects are mediated through the activation of the p44/42 MAP kinase, PI-3 kinase and focal adhesion kinase (FAK) pathways, as inlubition of the MAP kinase and PI-3 kinase pathways by PD98059 and wortmannin blocked progranulin-induced cell migration, and progranulin was found to hyper-phosphorylate FAK in endothelial cells. The association of strong progranulin expression with highly proliferative epithelial cells suggests that progranulin may take part in epithelial homeostasis. To investigate this hypothesis, we either overexpressed or downregulated the progranulin gene in SW-13 and MDCK cells. SW-13 cells are derived from human adrenal cancer, but their growth resembles non-transformed epithel

In this thesis we examine the effect of hormone replacement therapy (HRT) on the risk of colorectal cancer. Also examined are the effects of oral versus transdermal estrogen replacement therapy, methods of defining estrogen exposure, selection bias, and trends in the use of HRT over the last two decades. / A nested case-control study was conducted using records from Saskatchewan Health's administrative databases. Information on covariates not available from the databases was collected during interviews, from a subgroup of subjects. Incidence density sampling was used to age match controls (four per case, N = 12,116) to each of 3,059 cases accrued in the province from 1981 to 1998. / Short and long durations of HRT use (<5 years and &ge;5 years) were associated with odds ratios (OR) of 0.86 (0.76--0.97) and 0.78 (95% CI: 0.64--0.86), respectively. Stratification according to history of having had a screening sigmoidoscopy did not eliminate the observed protective effect. Important differences were not seen between more recent HRT use (<5 and <10 years), compared with more distant past use (&ge;5 and &ge;10 years). / The use of various definitions of estrogen exposure produced ORs ranging from 0.78 to 0.99 which are similar to results from almost two dozen observational studies conducted over the past two decades indicating that this is an important source of variability that needs to be considered. / The study of independent effects of oral and transdermal estrogens revealed a protective effect of transdermal estrogen that was much greater than that of oral estrogen and which has not previously been reported. A protective effect remained when women who had used oral estrogen only were used as the reference group. / Data pertaining to lifestyle factors collected by interview appeared not to alter ORs for HRT and colorectal cancer. However, due to extremely low response rates in the interview phase of the study, 30% among cases and 18% among controls, we were unable to conclude whether or not confounding was eliminated. / An important finding of research is the strong observed protective effect of transdermal estrogen replacement therapy. This demonstrates the importance of taking into consideration the mode of estrogen delivery in studies where the associations between HRT use and health outcomes are examined.