UCSF

Breast cancer is the most common non-skin cancer in women and the second leading cause of cancer death in women in the United States. Effective chemotherapy can potentially treat micrometastases as well as the primary tumor, reducing the risk of metastasis later in the lifespan and improving overall survival. Chemotherapy implemented prior to surgery (neoadjuvant chemotherapy) allows for the additional benefits of 1) potential tumor down-staging and the option of breast-conserving surgical methods and 2) the ability to assess the tumor's response to chemotherapy while the tumor is still present. Currently, it is difficult to predict tumor response to chemotherapy and effective methods of monitoring tumor response to treatment are needed.

Diffusion-weighted magnetic resonance imaging (DW-MRI) is sensitive to the random motion of water molecules, allowing for macroscopic detection of microscopic changes in cell density and cell content. This research focused on the use of DW-MRI in the monitoring of diffusion within the tumor during the course of treatment with neoadjuvant chemotherapy. In retrospective studies, the ability of DW-MRI measurements to predict patient outcomes was assessed. Problems related to the DW-MRI acquisition were identified and a technical solution was optimized for use in patients with breast cancer. Methods for improved DW-MRI processing were developed. Promising technical and processing innovations were evaluated in prospective clinical studies.

In retrospective studies, results from limited patient numbers suggested that DW-MRI measurements may predict long-term patient outcomes. Multiple barriers to DW-MRI analysis were identified, including image quality and the time needed for manual delineation of tumor regions of interest. Use of a new sequence resulted in improved image quality, quantitative differences in measurements related to the tumor apparent diffusion coefficient (ADC) distribution, and could allow for diffusion to be more accurately monitored throughout treatment. New segmentation methods were developed and could improve the efficiency of DW-MRI processing.

In this work, diffusion was identified as a marker of long-term patient outcomes and developments in image acquisition and image processing showed promise in further improving the ability to measure tumor diffusivity and monitor changes throughout treatment with neoadjuvant chemotherapy. Validation in large, prospective studies is needed.