Automated Quantification of Coronary Artery Calcifications on Radiotherapy Planning CTs for Cardiovascular Risk Prediction in Breast Cancer Patients: the BRAGATSTON Study

Breast Cancer, Cardiovascular Diseases

Breast cancer, Cardiovascular diseases, Coronary artery calcium, Prediction, Risk stratification, Deep learning technique, Software development, Computed tomography scan

The aim of the BRAGATSTON study is to provide a low cost tool for measuring CAC in breast cancer patients, thereby identifying patients at increased risk of CVD. Breast cancer patients and doctors can act upon this, by adapting the treatment and/or by adopting cardioprotective interventions. Hereby, the burden of CVD in breast cancer survivors can be reduced and better overall survival rates can be achieved.

Cardiovascular disease (CVD) is the second most common cause of death in breast cancer patients. Certain adjuvant treatments (e.g. anthracyclines, radiotherapy) increase the risk of CVD, in particular in patients with pre-existing CVD risk factors. Early identification of these patients enables targeted cardioprotective interventions, and switching to less cardiotoxic treatments. The strongest independent CVD risk factor is the presence and amount of coronary artery calcium (CAC). In clinical practice, CAC is quantified on cardiac CTs. Breast cancer patients treated with radiotherapy routinely undergo planning CTs. These inferior quality CTs may give the opportunity to routinely assess CAC. The BRAGATSTON study is a multicenter retrospective observational cohort study (UMC Utrecht, Erasmus Medical Center and Radboudumc). The project is divided into three work packages (WP), each WP with a unique aim: WP 1: To develop/optimize and validate the UMC Utrecht developed automated software to determine the presence and the amount of CAC (CAC score) on radiotherapy planning CT scans of breast cancer patients. WP 2a: To evaluate the association between CAC score measured automatically on planning CT scans and the risk of incident (non-)fatal cardiovascular events in breast cancer patients. WP 2b: To evaluate if the association as defined by work package 2a is modified by type of adjuvant treatment. WP 2c: To evaluate the association between other candidate imaging biomarkers (e.g. calcifications in other structures like the aorta or heart valves, the amount and distribution of body fat, bone characteristics) measured (automatically) on planning CT scans and the risk of incident (non-)fatal (cardiovascular) events in breast cancer patients. WP 3: To assess the added value of CAC score measured automatically on planning CT scans over classic cardiovascular risk factors and over treatment characteristics to predict (non-)fatal cardiovascular events in breast cancer patients.

None listed

Netherlands