Grant: Doctoral Award - Frederick Banting and Charles Best Canada Graduate Scholarships

The major goal of my research is to personalize cancer medicine by applying computational models to big data. Each year, ~1.4 million men around the word are diagnosed with prostate cancer and ~290,000 die from the disease. Localized cases are classified as low-risk, intermediate-risk or high-risk using clinical measurements to help guide treatment decisions. However, these groupings are not precise: for example 30-50% of intermediate-risk patients suffer relapse of their disease after radiotherapy or surgery. One reason for differing response to treatment is the cellular and molecular heterogeneity of tumours (i.e. not all tumours are the same). Tumour cells also exist within an environment that has gradients of oxygen levels. Patients with hypoxic tumours (i.e. tumours with low oxygen) can rapidly fail treatment. Focusing on 548 prostate cancer samples, I have shown that important genes (e.x. PTEN and RNLS) have an increased frequency of having a DNA copy number change under hypoxia. We also see more point mutations in important genes (e.x. TP53) in samples that were hypoxic. My goal is to further link tumour genomics, microenvironment and response to treatment. I will relate hypoxia to mutational trinucleotide signatures and epigenetic changes. I will also examine fusion genes and changes in splicing under hypoxia. Next, I will spend four months at the Wedge lab (Oxford University, UK) to look at how tumour evolution changes under hypoxia. Finally, I will use machine-learning to link these data and create a tool that can be used in the clinic to guide more precise treatment of prostate cancer. With the computational facilities at the Ontario Institute for Cancer Research, I am well equipped to do this work. Better understanding this relationship is critical to improving the lives of the ~25,000 Canadian men diagnosed with prostate cancer each year.