Image Driven Multi-Scale Modeling to Predict Treatment Response in Breast Cancer

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): The ability to identify-early in the course of therapy-patients that are not responding to a particular neoadjuvant regimen would provide the opportunity to switch to a potentially more efficacious treatment and transform current practice. Unfortunately, existing methods of determining early response are inadequate. The vision for this program is to develop tumor-forecasting methods for predicting response in individual breast cancer patients after a single cycle of neoadjuvant therapy. We propose to combine time-resolved drug- response cell scale data with physiological and tissue scale imaging data in order to initialize and constrain a multi-scale angiogenesis-cell proliferation model designed to predict both size and spatial characteristics of breast tumors at the completion of therapy. To achieve this goal, we will pursue the following specific aims: 1. (Pre-clinical validation) In the BT-474 HER2+ human breast cancer cell line, we will obtain: 1a. (cell scale) in vitro data quantifying rates of entry of proliferating cells into quiescence and apoptosis; 1b. (physiologica scale) in vivo MRI and PET measurements of cellularity, vascularity, and metabolism; 1c. (tissue scale) in vivo MR elastography measurements to quantify the tumor mechanical properties; 1d. (all scales) in situ data from fixed tumor tissue to corroborate cell and imaging-based metrics. These data will be integrated into the multi-scale model to predict tumor response after one cycle of the targeted anti-HER2 agents trastuzumab and lapatinib. 2. (Clinical application) In HER2+ patients receiving neoadjuvant trastuzumab and lapatinib, we will obtain: 2a. (physiological scale) in vivo MRI and PET measurements of cellularity, vascularity, and metabolism; 2b. (tissue scale) in vivo MR elastography measurements to quantify tumor mechanical properties. Guided by the results from Aim 1, these data will be integrated into the multi-scale model and make predictions on breast tumor response outcomes after a single cycle of trastuzumab and/or lapatinib. If successful, our approach would be the foundation for high-impact, large-scale application in clinical settings.