Abstract #2217

Classification and Prediction of Prognostic Factors of Breast Cancer Patients by MR Metabolomics

Guro Fannelb Giskedegrd¹, Steinar Lundgren^1,2, Beathe Sitter¹, Hans Fjsne³, Jostein Halgunset⁴, David E. Axelson⁵, Ingrid Susann Gribbestad¹, Tone Frost Bathen¹

¹Dept. of Circulation and Medical Imaging, Norwegian University of Science and Technology (NTNU), Trondheim, Norway; ²Dept. of Oncology, St. Olavs University Hospital, Trondheim, Norway; ³Dept. of Surgery, St. Olavs University Hospital, Trondheim, Norway; ⁴Dept. of Laboratory Medicine, Children's and Women's Health, NTNU, Trondheim, Norway; ⁵MRi_Consulting, Kingston, Canada

Predicting prognostic factors of breast cancer is important for clinical decision-making. The purpose of this study was to predict lymphatic spread and ER status of breast cancer patients using MR metabolomics, and to classify the patients according to these factors. HR MAS MR spectra of tumor tissue from breast cancer patients were obtained and further analysed by PLS and BBN. PLS analysis clustered the spectra according to lymphatic spread and ER status, and both PLS and BBN could predict the status of spectra from blind samples. MR metabolomics may thus be a tool to identify subclasses of breast cancer patients related to prognosis and outcome.