Aline Talhouk

High-grade serous ovarian cancer (HGSOC) represents approximately 70% of cases of ovarian carcinoma. Investigation into its molecular features led to the discovery of four subtypes of HGSOC, which differ from each other in their gene expression, cellular composition and survival outcomes. However, methods of whole-genome expression clustering have limited the reproducibility of existing subtype-specific research findings as well as being unsuitable for use in individual cases, such as in a clinical setting. PrOTYPE has been developed as a clinical-grade tool for determining the subtype of HGSOC tumours according to expression of a select panel of genes. To investigate the potential utility of determining HGSOC subtype using PrOTYPE in the clinical decision-making process, this thesis has two aims: 1) explore subtype-specific differences in survival outcomes of HGSOC patients treated with the anti-angiogenic drug bevacizumab, to determine whether subtype could be used to inform treatment decisions, and 2) evaluate the subtypes of paired samples collected before and after treatment to platinum and taxane chemotherapy to determine whether exposure to chemotherapy induces changes in subtype. In Aim 1, I found that cases belonging to the immunoreactive C2.IMM subtype generally had improved overall and progression-free survival when treated with bevacizumab. Results also suggested that the C5.PRO subtype may be associated with poorer survival when treated with bevacizumab compared to if treated with platinum and taxane chemotherapy alone. However, these findings could not be validated in independent cohorts; stratification of patients by subtype-specific treatment response in a clinical setting would require further investigation before being implemented. Additionally, a search for novel genes associated with differential treatment outcomes with bevacizumab identified several genes that may be useful in future research on targeted approaches to bevacizumab treatment. In Aim 2, I found that change in subtype from before to after exposure to neoadjuvant chemotherapy occurred in the majority (11/16) of the cases studied; this indicates that pre-chemotherapy subtype cannot be determined from post-chemotherapy samples. Clinical and prognostic significance of HGSOC subtype after chemotherapy remains an area for future investigation. Subtyping using PrOTYPE could be used in the clinical context to guide treatment decisions and evaluate patient prognosis.

View record

The full abstract for this thesis is available in the body of the thesis, and will be available when the embargo expires.

View record

This thesis presents two contributions. The first contribution deals with the problem of siloed data collection and prohibitive data acquisition costs. These costs limit the size and diversity of datasets used in health research. Access to larger and more diverse datasets improves the understanding of disease heterogeneity and facilitates inference of relationships between surgical and pathological findings with symptomatic indicators and outcomes. Unfortunately, freely enabling access to these datasets has the potential of leaking private information, such as medical records, even when these datasets have been stripped of personally identifiable information.In the first part of this thesis, we present LEAP, a data analytics platform with support for federated learning. LEAP allows users to analyze data distributed across multiple institutions in a private and secure manner, without leaking sensitive patient information. LEAP achieves this through an infrastructure that maintains privacy by design and brings the computation to the data, instead of bringing the data to the computation. LEAP adds an overhead of up to 2.5X, training Resnet-18 with 15 participating sites, when compared to a centralized model. Despite this overhead, LEAP achieves convergence of the model’s accuracy within 20% of the time taken for the centralized model to converge.One of the techniques used by LEAP to preserve the privacy of sensitive queries is differential privacy. Successive DP queries to a dataset depletes the privacy budget. When the privacy budget is depleted, data curators must block access to the underlying dataset to prevent private information from leaking. In the second part of this thesis, we present a system called the SmartCache. The SmartCache optimizes the use of the privacy budget by interpolating old query results to help answer new queries using a synthetic dataset. Queries answered from the synthetic dataset have a smaller privacy cost, so more queries can be answered before the budget runs out. For statistical queries, the SmartCache saved 30%-50% of the budget for threshold values of 0.99 and 0.999, and for gradient queries it consumed 70% less of the privacy budget when training a fully connected model.

View record