The work proposed under this continuation proposal is designed to support the development, optimization and implementation of quantitative MR imaging markers for assessing response to treatment for primary breast cancer. In the initial project period, we developed a breast MRI method that employs a three-time point analysis of contrast kinetics, the signal enhancement ratio (SER), to assess tumor vasculature, while providing high spatial resolution to assess lesion morphology. We evaluated the effectiveness of the SER method for breast cancer staging and also applied the technique to assess the response of primary breast cancers to pre-operative, or neoadjuvant chemotherapy. Preliminary results from the neoadjuvant MRI studies suggested that the accuracy of MRI for staging tumor extent and measuring size changes in response to treatment resulted in better prediction of recurrence-free survival. Based upon these results, the multi-center ACRIN trial 6657, evaluating CE-MRI for assessment of patients undergoing neoadjuvant chemotherapy for locally-advanced breast cancer, opened in May 2002 and is nearing its accrual target of 244 subjects. ACRIN 6657 is an observational trial evaluating MRI's ability to identify patient subgroups with statistically-significant differences in survival outcome, that could not be identified on the basis of standard clinical and radiographic assessment alone. If this can be demonstrated, it may be possible that subsequent neoadjuvant trial designs can introduce patient triaging based upon MRI measurements. The goal of this continuation proposal is to further improve and evaluate the effectiveness of quantitative MRI measures for assessing primary breast tumors, to support their use as in-vivo predictive markers that can be used to guide treatment. The three aims address the data acquisition protocol (Aim 1), the optimization of individual predictor variables (Aim 2), and the development of predictive models to optimize linear combinations of imaging markers (Aim 3). These developments are proposed in concert with the separate development of a multi-center clinical trial protocol in which these techniques will be tested for their ability to measure early tumor changes in response to treatment and predict overall treatment response.
Under Aim 1, the image acquisition protocol will be advanced to use current state-of-the art technology to maximize the combined morphologic/functional characterization of breast tumors. Three additional functional measurements, tumor ADC measured by diffusion-weighted MRI, tumor choline concentrations [tCho] measured by 1H single voxel MR spectroscopy, and tumor perfusion measured by arterial spin labeling (ASL) MRI, will be added to the breast MR exam.
Aim 2 will use existing data from the UCSF pilot study and ACRIN 6657 to optimize individual predictor variables by investigating the effect of threshold settings on predictive performance.
Aim 3 will apply new statistical approaches for optimally combining markers in a predictive model. ? ? ?
Showing the most recent 10 out of 26 publications
