Because of its excellent soft tissue contrast, Magnetic Resonance Imaging (MRI) may aid early detection and staging of breast and prostate cancer. MRI has high sensitivity but its specificity has been disappointing. New methods that improve identification of metastatic lesions by MRI would have a significant impact on early detection and treatment of cancer.

Previous work in this laboratory showed that high spectral and spatial resolution (HiSS) MRI improves image quality and detection of the effects of contrast agents. HiSS images can be acquired with clinically acceptable run times by using frequency resolved echo planar methods to obtain detailed spectra of the water and fat resonances associated with each image voxel. The proposed research tests the hypothesis that contrast enhanced HiSS MRI can distinguish metastatic tumors from non-metastatic tumors based on measures of image texture and edge delineation. We will develop methods for acquiring and processing HiSS datasets to maximize contrast, edge delineation and signal-to-noise ratio. Contrast media will be used to enhance contrast, improve depiction of tumor boundaries, and detect tumor vascular structure.

We will test these experimental MRI methods using the well-characterized Dunning system of prostatic cancers. We propose to use four cell lines that vary in metastatic ability. HiSS MRI data will be correlated with tumor growth rate and size, microvessel density, degree and spatial distribution of necrosis, and number of overt metastases. This will improve our understanding of the biological processes that determine contrast in HiSS images and optimize HiSS MRI to detect features that differentiate metastatic and non-metastatic cancers. We will determine whether properties of the primary tumor measured by MRI can predict the behavior of metastases.

The specific areas of innovation are: 1) Use of HiSS for anatomic and functional imaging 2) Application of HiSS to differentiate between metastatic and non-metastatic cancers 3) Application of CAD (computer aided diagnosis) methods developed by Dr. Giger and coworkers to analyze HiSS data 4) Use of molecular biologic methods to produce model tumors that guide the development of new imaging methods. The proposed research brings together a strong interdisciplinary team including Dr. Giger (image analysis, CAD), Dr. Rinker-Schaeffer (molecular biology, cancer biology), and Dr. Karczmar (MRI, tumor physiology). This work would have a significant and rapid impact on clinical practice.

R21CA089408

2002-04-01

2005-03-31