Radonc/BME Medical Image and
Computational Analysis Laboratory
Simeng Zhu
2nd Medical Student
UF College of Medicine

email link
  1. Resume/CV (pdf)
  2. Research Overview

    Modeling Patterns of Breast Cancer Metastases

    Background and Significance
    In 2015, ~232,000 women in the US will be diagnosed with invasive breast cancer.2,3 Approximately 11% (26,000) of these will develop distant metastases only after an initial M0 diagnosis (no metastases)4 and will account for nearly 2/3 of the 40,000 breast cancer deaths annually.3 Current National Cancer Care Network (NCCN) guidelines for asymptomatic breast cancer omit proactive imaging for the detection of early-stage metastatic disease, regardless of a patient’s risk.5 Thus, when detected because of an outward symptom, such as persistent cough, bone pain, headaches or dizziness, metastatic disease is typically advanced, with multiple large masses afflicting multiple organs. Under these guidelines, metastatic breast cancer carries a discouraging 20% overall survival (OS) and 2% disease-free survival (DFS) at 5 years.6 However, there is a subset of women for whom systemic therapy coupled with early detection and local treatment of the residual metastases achieves a marked improvement in survival, yet the approach to staging metastases in breast cancer has not evolved substantially over the past 3 decades.

    Our long-range, overall hypothesis is that by adding surveillance imaging to standard follow-up care for high-risk breast cancer survivors a meaningfully large subset of patients will achieve dramatically increased overall survival (OS) and disease-free survival (DFS). Our immediate objectives are to gather additional data and better model the patterns of early metastatic spread of breast cancer to provide additional evidence of the efficacy of our approach and to optimize the surveillance imaging protocol to strengthen subsequent applications to the NIH, DOD Breast Cancer Research Program, the Komen Foundation, and/or other external agency.

    We will conduct a retrospective outcomes analysis of patients treated for metastatic breast cancer at UF, based upon the data collection and analysis methods presented recently by our collaborator Dr. Cristiane Takita and her clinical team at the University of Miami. We will retrospectively review the records of all patients treated with SBRT or SRS at UF from 2010-2014 for metastatic breast cancer. Oligometastatic state will be determined by imaging and clinical documentation. We will record the initial stage of disease (I – IV), and tumor ER, PR, and HER-2 status. Kaplan Meier statistics will be used to determine survival after metastasis detection. The Cox proportional hazard model will be used to assess the effect of patient, tumor, and treatment characteristics as predictors of survival after metastases. This emergent UF data will be analyzed independently and then combined with the UM data to strengthen their sub-type analysis.
  3. Related MIACALab projects
    1. Tumor early detection in the lung and brain.
    2. Lung radiation dose response

  4. Close Research Collaborators
    1. W. O'Dell (Radiation Oncology Research)
    2. Dr. Julie Bradley
    3. Dr. Cristiane Takita