UCSF

Radiation therapy is one of the most common curative therapies for patients with localized prostate cancer, but despite excellent success rates, a significant number of patients suffer post- treatment cancer recurrence. The accurate characterization of early tumor response remains a major challenge for the clinical management of these patients. Multi-parametric MRI/1H MR spectroscopy imaging (MRSI) has been shown to increase the diagnostic performance in evaluating the effectiveness of radiation therapy. 1H MRSI can detect altered metabolic profiles in cancerous tissue. In this project, the concentrations of prostate metabolites from snap-frozen biopsies of recurrent cancer after failed radiation therapy were correlated with histopathological findings to identify quantitative biomarkers that predict for residual aggressive versus indolent cancer. The total choline to creatine ratio was significantly higher in recurrent aggressive versus indolent cancer, suggesting that use of a higher threshold tCho/Cr ratio in future in vivo 1H MRSI studies could improve the selection and therapeutic planning for patients after failed radiation therapy.

Varying radiation doses may cause a diverse effect on prostate cancer micro-environment and metabolism, which could hold the key to improving treatment protocols for individual patients. The recent development and clinical translation of hyperpolarized 13C MRI have provided the ability to monitor both changes in the tumor micro-environment and its metabolism using a multi-probe approach, [1-13C]pyruvate and 13C urea, combined with 1H Multi-parametric MRI. In this thesis, hyperpolarized 13C MRI, 1H dynamic contrast enhancement, and diffusion weighted imaging were used to identify early radiation dose response in a transgenic prostate cancer model. Hyperpolarized pyruvate to lactate metabolism significantly decreased in a dose dependent fashion by 1 day after radiation therapy, prior to any changes observed using 1H DCE and diffusion weighted imaging. Hyperpolarized 13C urea and 1H DCE both show increase in perfusion/permeability by 4 days post-radiation. In tumor region treated with high dose radiation, ADC values significantly increased post-radiation, suggesting a decrease in cellular density. These dose dependent changes can be used as markers of early tumor response to the impact of increasing doses of radiation therapy. In addition, a spectral-spatial pulse sequence was developed for the 14T to dynamically observe kinetic information in a transgenic prostate cancer model before and after radiation therapy. A novel modeling approach was proposed to parameterize perfusion in the kinetic modeling of pyruvate to lactate conversion for better characterization of pyruvate metabolism. Unlike single time point HP 13C urea imaging, quantitative pharmacokinetic parameters such as blood flow and extracellular extravascular volume fraction can be extracted from dynamic acquisitions. Blood flow measured by hyperpolarized 13C urea was highly correlated with Ktrans measured by 1H DCE, suggesting hyperpolarized urea might be able to provide similar information as 1H DCE.

The results of this thesis show that Multi-parametric MRI, including functional MRI, 1H MRSI, and hyperpolarized 13C, holds great potential for evaluating early tumor response to radiation therapy of prostate cancer. The findings of this thesis will be useful in designing future studies for using combined Multi-parametric 1H and hyperpolarized 13C MRI to improve planning and assessing radiation therapy in individual prostate cancer patients.