High incidence of fractures is reported in women affected by breast cancer undergoing hormone deprivation therapy (HDT). To date, the available tools for the identification of skeletal fragility in this clinical setting are not accurate. Goal of the study is to identify CT-based radiomics features able to predict vertebral fractures (VFs) in breast cancer women exposed to HDTs, and to generate a scoring system associated with the highest VF risk, as defined by a machine learning framework that includes clinical and laboratory data, dual-energy X-ray absorptiometry (DXA) parameters and radiomics features.

With this aim, the project will combine a cross-sectional and a prospective study in subjects exposed to HDTs who will be evaluated for skeletal fragility. The results of this study might help clinicians in identifying cancer patients to be treated with bone-active drugs in primary prevention and could be useful for improving prescription appropriateness in osteoporosis induced by HDTs.

The project involves the Humanitas Research Hospital and the Department of Electronics, Information and Bioengineering of the Politecnico di Milano.