Grant: DRA : Canada Graduate Research Scholarship - Doctoral Program (CGRS D)

Ovarian cancer is one of the most deadly cancers affecting women, with many patients diagnosed at an advanced stage (nearly 3,000 cases were registered in Canada alone in 2024). Even with surgery and chemotherapy, most women experience the cancer coming back, leading to poor long-term survival. To improve outcomes, a targeted drug called bevacizumab has been added to treatment in some patients. This drug works by blocking the blood supply that tumors need to grow, and it has been shown to delay the return of cancer across multiple large-scale international clinical trials. However, bevacizumab is very expensive, can cause serious side effects, and unfortunately, does not work for every patient. Right now, doctors do not have a reliable way to predict who will benefit from this treatment. As a result, many women may receive a drug that will not help them, while also facing the risks and costs associated with it. Our research aims to address this challenge by using artificial intelligence (AI) to study digital images of tumor tissue, which are already collected during diagnosis as a routine practice. By training advanced computer models to recognize patterns in these images, we aim to identify new "biomarkers" that can tell us which patients are most likely to respond well to bevacizumab. We will also combine these image patterns with genetic and clinical information to build stronger, more accurate tools for prediction. The impact of this work is two-fold: patients who are most likely to benefit from bevacizumab could be identified early and receive the drug, while those unlikely to benefit could be spared unnecessary side effects and costs. Beyond ovarian cancer, this approach could transform how we match patients to the right treatments in many other cancers, improving personal health and saving healthcare resources with more personalized and effective care.