Cancer includes a wide range of diseases that can affect any part of the body characterized by the rapid growth of abnormal cells. Based on the World Health Organization, cancer is a leading cause of death worldwide, accounting for nearly 10 million deaths in 2020, or almost one in six deaths. The significance of early cancer diagnosis cannot be overstated. It often makes the difference between successful treatment and a challenging prognosis. Efficient algorithms supporting cancer research, diagnosis, and treatment are essential to enable timely detection and make informed decisions about the risks and benefits of different treatment options.

The development of two tools is planned in the project: CanAge and CanSurv. CanAge leverages genetic data to predict when a specific cancer might develop, offering personalized insights into germline genomic contributions to cancer risk. It may also serve as a foundation for future integrated scores considering both genomic and environmental factors, helping identify individuals with increased cancer development risks. CanSurv is designed to be a multivariable risk prediction model capable of estimating cancer survival for individual patients and assessing the incremental benefits of any adjuvant therapies for which clinical trial data are available. CanSurv is a publicly available decision support tool that empowers both patients and physicians to make well-informed shared decisions regarding the risks and benefits of cancer treatment.

The leader of the project titled "Advanced statistical and deep learning models to predict cancer risk and support treatment decisions" financed by the National Science Center under the SONATA BIS 13 program is Michal Marczyk, PhD, DSc. As part of the project, it is planned to create a new research team by employing a full-time researcher in a post-doc position and two students or doctoral students employed under the scholarship. The tools developed in this project could have a huge impact on personalized medicine in cancer treatment.