Evaluation of predictive models on the occurrence of breast cancer at 5 years

This project aimed to evaluate and improve cancer risk prediction models using CARTaGENE data. The initial objective was to assess the performance of the Breast Cancer Risk Assessment Tool (BCRAT), a widely used clinical model developed in the United States, and to determine whether adding polygenic risk scores derived from genetic data could improve the prediction of breast cancer occurrence within five years among women in Quebec. By combining health, lifestyle, family history, and genetic information available in CARTaGENE, the project evaluated how well these tools identify individuals at increased risk of developing cancer. The research also led to the development of new statistical methods for assessing and comparing risk prediction models and genetic risk scores. Subsequent analyses extended this work to other cancers, including lung cancer, where researchers evaluated the performance of risk-based screening approaches and identified novel genetic variants associated with lung cancer susceptibility. Overall, the project contributed to improving the evaluation, calibration, and application of personalized cancer risk prediction tools in the Quebec population and supported the development of more targeted prevention and screening strategies.

• Cancer
• Genetics and genomics
• Methodology and biostatistics

• Genetic data
• Linked data
• Questionnaire data