Epidemiological risk model to inform targeted prevention strategies in endometrial cancer
Principal Investigator: Dr. Aline Talhouk
Affiliation: University of British Columbia
Start Year: 2020
We will be using two risk models to predict the incidence of endometrial cancer and its precursor lesion, endometrial hyperplasia, in a Canadian population. These pre-existing models have both resulted in predicting a woman’s absolute risk of developing invasive endometrial cancer and can help to identify women who are at high-risk. This allows for targeted interventions to promote prevention along with early detection of cancer. By using the CanPath data, we will validate and improve these two pre-existing risk models by further exploring how additional factors, such as socioeconomics, ethnicity, gender variations and environmental exposures, may also contribute to endometrial cancer risk. Other health conditions are known to contribute to endometrial cancer, but their causal relationships have not been fully explored. Obesity, polycystic ovarian syndrome, and an increased exposure to female hormones are associated with higher incidence of endometrial cancer. Using data collected by CanPath, we will conduct our research using statistical and machine learning techniques. Machine learning is an application of artificial intelligence that enables us to analyze big data along with traditional epidemiological approaches. The purpose of creating an enhanced risk model is to reverse the current alarming trends in both incidence and mortality for endometrial cancer and to better understand health disparities in Canada. Our hope is that this model will identify high-risk women in Canada and recommend personalized prevention strategies based on a woman’s risk. These prevention strategies may include lifestyle modification, medication, and surgical prevention.