Hybrid, multidimensional magnetic resonance imaging of prostate cancer

Abstract

Prostate cancer is the most commonly diagnosed cancer among men, estimated to affect one in six American men. Despite its high prevalence, nearly 50% of prostate cancer is insignificant and unrelated to cause of death of the patient. It is thus of paramount importance that physicians are able to make informed decisions on course of treatment for prostate cancer patients; an aggressive treatment option such as radiation therapy or surgery with risk of adverse side effects is necessary for high grade cancers while active surveillance may be sufficient for lower grade cancers. Current methods for assessing the aggressiveness of prostate cancer are insufficient and repeat biopsies or histological analysis of prostatectomy specimens often show disagreement with initial staging. MRI has become a useful, non-invasive tool for prostate cancer screening. A clinical multiparametric MRI (mp-MRI) exam typically includes T1, T2 and apparent diffusion coefficient (ADC) maps, extracted from T1-weighted, T2-weighted and diffusion-weighted MRI (DWI) sequences, respectively. However, the efficacy of PCa MRI is limited by the inability to distinguish the signal from cancerous foci from the background signal from normal tissue using standard, clinical MRI sequences. This dissertation investigates a hybrid, multidimensional imaging approach to prostate MRI, which looks at how standard measures (ADC, T1, T2) react to changes in sequence parameters (TE, TR, b). This enables a selective filtration of particular tissue components and subsequent emphasis of differences in tissue structure indicative of prostate cancer. My results suggest that using hybrid imaging to identify differences in tissue structure could lead to better differentiation between normal tissue, aggressive prostate cancer and benign tumors or conditions. Hybrid imaging provides structural information undetectable using conventional mp-MRI. It may therefore provide independent, diagnostic information that compliments ADC, T1 or T2 values and increase diagnostic accuracy when used in combination with conventional mp-MRI. Structural markers such as decreased luminal volume, increased cellularity, and nucleomegaly become more prominent with increased Gleason Score. Hybrid MRI may therefore be useful for noninvasively determining Gleason Score in the future.