Grant: Project Grant

Blood clots are a known complication of cancer and its treatment, and may lead to serious short- and long-term complications including death. In children, blood clots are often linked to a central venous line, a device inserted in the body to deliver medicine, fluids, or nutrients. While effective options exist to prevent blood clots, they are associated with side effects, costs, and increased burden on patients. No test or algorithm is currently available to help healthcare professionals identify which children are more likely to develop blood clots. Children at high risk of blood clots could benefit from medicine to treat adequately blood clots and minimize their impact. We will perform a study enrolling children newly diagnosed with cancer (solid cancer outside the central nervous system) requiring chemotherapy or radiotherapy and a central venous line. Bloodwork will be performed on 4 to 5 occasions to identify blood tests, called biomarkers, that could predict blood clots. We will review health records and perform painless tests called Doppler ultrasounds to detect blood clots in participants. We will follow the children for 18 months to determine whether they develop blood clots and complications of blood clots including catheter-related complications, post-thrombotic syndrome, and occlusion of blood vessels. A risk-assessment model taking into account clinical variables and laboratory results will be created to allow early identification of children at high risk of blood clots. We will use traditional statistical methods and machine learning to build this risk-assessment model. As survival from childhood cancer has increased drastically in the past decades, it is increasingly important to prevent acute complications such as blood clots and reduce late effects of treatment.