This thesis employed methods used in health care decision making to evaluate strategies for prevention and treatment of febrile neutropenia (FN) in non-Hodgkin lymphoma (NHL) patients. The objectives of this thesis were to quantify the cost-effectiveness of filgrastim and pegfilgrastim as primary prophylaxis against FN in NHL patients, to develop an algorithm for converting health-related quality of life data collected in non-Hodgkin lymphoma patients into preference-based health utility values, and to evaluate NHL patients’ preferences for outpatient treatment of FN. The cost-effectiveness analysis demonstrated that neither filgrastim, nor pegfilgrastim are cost-effective, with respective incremental cost-effectiveness ratios [95% confidence interval] of $4,599,000/QALY [$597,045, dominated] and $6,272,000/QALY [$730,692, dominated], well above the normally accepted threshold of $50,000/QALY. The algorithm for deriving health utility values was based on a regression model that used health utility values obtained from the EQ-5D instrument as the outcome variable and the four subscales of the Functional Assessment of Cancer Therapy – General (FACT-G) questionnaire as the predictor variables. The model final model included three of the FACT-G subscales, and had an R-squared value of 0.502 and a mean squared error of 0.013. A discrete choice experiment was used to examine patients’ preferences for out patient treatment of FN, and demonstrated that out-of-pocket costs, unpaid caregiver time required daily, and probability of return to hospital are all significant attributes when considering outpatient therapy for FN. Adjusted odds ratios [95% confidence intervals] of accepting outpatient treatment for FN were 0.84 [0.75 to 0.95] for each $10 increase in out-of-pocket cost; 0.82 [0.68 to 0.99] for each 1 hour increase in daily unpaid caregiver time; and 0.53 [0.50 to 0.57] for each 5% increase in probability of return to hospital. These results provide important information for clinicians and health care decision makers involved in implementing programs for NHL patients with FN.

The ability of cancer cells to escape anti-tumor immune responses is acknowledged as one of the hallmarks of cancer. Overexpression of immunoregulatory molecules is one mechanism responsible for the immunsuppressive network that is characteristic of the tumor microenvironment. In this thesis, we investigated the role of CD200, a potent immunoregulatory molecule, in Chronic Lymphocytic Leukemia. We showed that functional blockade of CD200 on lymphoma cells or primary CLL cells, both of which express CD200 at high levels, augmented cytotoxic killing of these cells by effector CD8+ T cells in vitro. We also identified and characterized a previously unrecognized soluble form of CD200, sCD200, present in elevated levels in CLL plasma when compared to plasma from controls. The data reported show that patients with high sCD200 levels have more aggressive disease, inferring that sCD200 may be a novel prognostic marker for CLL. The in vivo function of sCD200 was investigated for its ability to support engraftment of CLL splenocytes in NOD.SCID mice. Infusion of sCD200hi CLL plasma, but not sCD200lo normal plasma, enhanced engraftment of CLL-splenocytes in vivo, an effect which was abrogated by depletion of sCD200 from CLL plasma. The prolonged engraftment of CLL cells seen in this model (>6 months) suggests these mice represent a useful pre-clinical model for drug screening. The effect of CD200 blockade was tested in this model, and was found to be as effective in eliminating engrafted CLL cells as rituximab. Investigation of the mechanisms leading to the release of sCD200 from CLL cells showed that sCD200 was produced following ectodomain shedding by ADAM proteases and MMPs. Results from studies reported in this thesis support the hypothesis that CD200 plays a major role in CLL biology, and suggests it may represent a novel therapeutic target for CLL.

Metastasis initiation is a complex process encompassing numerous steps. To identify molecular alterations associated with early and late stages of metastasis, we used high throughput screening techniques. Early events in metastasis were investigated by differential proteomic analysis of lymph node-negative and positive breast cancer samples. Two candidate biomarkers (DCN and HSP90B1) were discovered and further validated through tissue microarray analyses. To examine late events in metastasis formation, we prospectively evaluated genomic differences between disseminated tumour cells in bone marrow, and metastatic tumour cells obtained from computed tomography guided biopsies of bone metastases. Results indicate that specific subsets of genes are required for breast cancer cells to initiate bone metastases. Discovery of proteomic and genomic alterations specifically associated with metastases may yield biomarkers capable of stratifying patients into different risk categories. Proteins and genes identified in this work may form the foundation of a biomarker panel for metastatic risk assessment.

Each cancer is unique: it reflects the underlying genetic make-up of the patient and the stochastic mutational processes that occur within the tumour. This uniqueness suggests that each patient should receive a personalized type of therapy. Current predictions of a cancer patient’s outcome or prognosis are highly inaccurate. To aid in the prediction of patient prognosis based on highthroughput molecular datasets I have worked to optimize each step of the experimental pipeline: platform annotation, experimental design, consideration of tumour heterogeneity, data pre-processing and statistical analysis, and feature selection. First, a 12k CpG Island clone library was sequenced and annotated using a BLAT analysis. Second, microarrays built using this library were used in a fully-saturated study to evaluate the importance of ChIP-chip experimental design parameters. Third, intra-tumour heterogeneity was shown to influence specific pathways in a large fraction of genes. Fourth, a systematic empirical evaluation of 19,446 combinations of microarray analysis methods identified key steps of the analysis process and provided insight into their optimization. Finally, the combination of a two-stage experimental design and a novel semi-supervised algorithm yielded a six-gene, mRNA abundance-based classifier that could divide non-small cell lung cancer patients into two groups with significantly different outcomes in four independent validation cohorts. Further, a permutation study showed that millions of six-gene markers exist, but that ours ranked amongst the top 99.98% of all six-gene markers. The knowledge gained from these studies provides a key foundation for the development of personalized therapies for cancer patients.

The ability of cancer cells to escape anti-tumor immune responses is acknowledged as one of the hallmarks of cancer. Overexpression of immunoregulatory molecules is one mechanism responsible for the immunsuppressive network that is characteristic of the tumor microenvironment. In this thesis, we investigated the role of CD200, a potent immunoregulatory molecule, in Chronic Lymphocytic Leukemia. We showed that functional blockade of CD200 on lymphoma cells or primary CLL cells, both of which express CD200 at high levels, augmented cytotoxic killing of these cells by effector CD8+ T cells in vitro. We also identified and characterized a previously unrecognized soluble form of CD200, sCD200, present in elevated levels in CLL plasma when compared to plasma from controls. The data reported show that patients with high sCD200 levels have more aggressive disease, inferring that sCD200 may be a novel prognostic marker for CLL. The in vivo function of sCD200 was investigated for its ability to support engraftment of CLL splenocytes in NOD.SCID mice. Infusion of sCD200hi CLL plasma, but not sCD200lo normal plasma, enhanced engraftment of CLL-splenocytes in vivo, an effect which was abrogated by depletion of sCD200 from CLL plasma. The prolonged engraftment of CLL cells seen in this model (>6 months) suggests these mice represent a useful pre-clinical model for drug screening. The effect of CD200 blockade was tested in this model, and was found to be as effective in eliminating engrafted CLL cells as rituximab. Investigation of the mechanisms leading to the release of sCD200 from CLL cells showed that sCD200 was produced following ectodomain shedding by ADAM proteases and MMPs. Results from studies reported in this thesis support the hypothesis that CD200 plays a major role in CLL biology, and suggests it may represent a novel therapeutic target for CLL.

Metastasis initiation is a complex process encompassing numerous steps. To identify molecular alterations associated with early and late stages of metastasis, we used high throughput screening techniques. Early events in metastasis were investigated by differential proteomic analysis of lymph node-negative and positive breast cancer samples. Two candidate biomarkers (DCN and HSP90B1) were discovered and further validated through tissue microarray analyses. To examine late events in metastasis formation, we prospectively evaluated genomic differences between disseminated tumour cells in bone marrow, and metastatic tumour cells obtained from computed tomography guided biopsies of bone metastases. Results indicate that specific subsets of genes are required for breast cancer cells to initiate bone metastases. Discovery of proteomic and genomic alterations specifically associated with metastases may yield biomarkers capable of stratifying patients into different risk categories. Proteins and genes identified in this work may form the foundation of a biomarker panel for metastatic risk assessment.

Haematopoiesis requires strict regulation in order to maintain a balanced production of the various blood cell components. Escape from this regulation contributes to the development of cancers such as leukemia. SOCS2 is a member of the Suppressor of cytokine signalling (SOCS) family, and normally functions as a negative regulator of the JAK/STAT pathway. I examined gene expression and promoter methylation in acute myeloid leukemia (AML) cell lines and patient samples. SOCS2 expression was quite variable in AML patients, and very low in acute promyelocytic leukemia (APL) patients. Promoter hyper-methylation was found in these patients, particularly those with high white blood cell count and a FLT3-ITD. I speculate that SOCS2 interacts with an aspect of the signalling complex to inhibit cell growth in these patients, and silencing SOCS2 is necessary for leukemia progression. Treating these patients with a de-methylating agent, such as decitabine, may show promise in the clinic.

Colorectal cancer (CRC) is common in Ontario. This study described the processes of care following CRC resection, and identified CRC relapse from administrative data. Methods: CRC patients aged 18-80 from 1996-2001 with a colorectal resection were identified from the Ontario Cancer Registry. Linked discharge abstracts and physician billings were examined for physician visits, body imaging and endoscopy over the 5 year follow-up period. Administrative codes suggesting disease relapse were compared with patient charts. Results: Overall, 12,804 patients were identified and 8,804 had no evidence of relapse. Most (96.2%) patients had general practitioner follow-up, while 49.3% had medical oncology and 80.4% had general surgery follow-up. Greater than 90% of patients received endoscopy, while only 68.7% of patients received body imaging. Detecting disease relapse was 87.5% sensitive and 93.0% specific. Conclusions: There is potential for improving post-resectional follow-up in CRC patients. It is possible to detect relapse through administrative databases.

Intensity modulated radiotherapy (IMRT) can potentially increase the dose delivered to liver tumours while sparing normal tissues from dose. More complex IMRT, with more modulation of the radiation beam is more susceptible to geometric and dosimetric uncertainties than simpler radiotherapy plans. Simple breath-hold liver IMRT using few radiation beam segments (<30) was investigated in 27 patients to determine the quality of treatment in terms of tumour dose coverage and normal tissue sparing as compared to index IMRT using >30 segments. In all 27 plans number of segments was reduced to <30 without compromising tumour coverage or normal tissue dose constraints, at the expense of dose conformity. Delivered tumour and normal tissue dose did not differ statistically between IMRT plans when accounting for treatment residual geometric error. This research supports considering the use of simple IMRT for treatment of liver cancer, except when loss of dose conformation is undesirable (i.e. very high doses).

Aberrant activation of the canonical Wnt signaling pathway accounts for the vast majority of colorectal cancers. The Rac1 GTPase is overexpressed in colon cancer, and its splice variant, Rac1b, is preferentially expressed in colon tumours. Rac1 and Rac1b have both been previously shown to crosstalk with the canonical Wnt signaling pathway in colon cancer; however, the specific means by which this crosstalk occurs were unclear. This study examines the molecular mechanisms of Rac1/1b in the cooperation with canonical Wnt signaling in colon cancer. In a colon cancer cell line with dysregulated Wnt signaling, the constitutively active Rac1 mutant, V12Rac1, was observed to transcriptionally upregulate the expression of a gene set associated with cellular migration. Further, V12Rac1-mediated promotion of cell migration was dependent on its nuclear localization. Previous work in our lab has shown a Rac1-specific activator, Tiam1, is present in the nucleus at the promoter of Wnt target genes upon Wnt3a stimulation; and that exogenous introduction of Tiam1 increased the expression of a Wnt-responsive reporter (TopFlash). Given the importance of nuclear localization of Rac1 in the promotion of tumourigenic processes, we demonstrated that knockdown of endogenous Tiam1 reduced TopFlash expression, proving reverse specificity and strengthening the evidence of a nuclear role for Rac1. Since some functional differences exist between Rac1 and Rac1b, we also examined Rac1b for transcriptional targets following induction, and identified the RhoA effector, ROCK2, which has been previously associated with cell migration. ROCK2 demonstrated a positive correlation with Rac1b transcript expression in primary colon tumours as compared to matched normal tissue specimens. Interestingly, the observed induction in ROCK2 transcript did not translate into a detectable change in protein expression or kinase activity. Like Rac1, Rac1b also promotes cellular motility, which is dependent on nuclear localization. Cell migration can be negatively regulated by E-cadherin. Following Rac1b knockdown in HT29 cells, we show that Rac1b might contribute to motility through upregulation of the E-cadherin-repressor, Slug. Taken together, we provide greater insight into the mechanistic roles of Rac1 and Rac1b in transcriptionally regulating target genes to promote cellular processes, such as cell migration, in colon cancer with dysregulated canonical Wnt signaling.