On the Use of MRI to Predict Individualized Treatment Strategies for Breast Cancer

• Activity Number: 506
• Type: Contributed
• Date/Time: Wednesday, August 6, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Section on Medical Devices and Diagnostics
• Author(s): Ying Lu*+ and John Kornak and David Newitt and Nola Hylton
• Companies: Palo Alto VA CSPCC/Stanford University and University of California, San Francisco and University of California, San Francisco and University of California, San Francisco
• Keywords: Medical Diagnosis ; Breast Cancer ; Decision Making ; MRI ; Optimization ; Survival Time

Abstract

New treatments are continually being developed for breast cancer. The additional treatment options available to the clinician beg the question of when should treatment be switched for a patient due to lack of treatment response? New work on treatment response as measured by MRI is based on percent signal enhancement (PE) and signal enhancement ratio (SER). The 'volume of enhancement' is defined as the region above PE/SER thresholds. Reduction in volume of enhancement is considered representative of treatment success. The threshold parameters for PE and SER can be chosen to optimize treatment response measurement. We therefore consider the threshold values as parameters to be optimized in the decision process. The optimization of thresholds and treatment choice is performed with respect to maximizing survival time (time to recurrence of breast cancer). We will present results based on a pilot data study and simulations of larger databases/studies.