UCLA

Prostate cancer is diagnosed in almost 200,000 men in the U.S. annually. With approximately 30,000 deaths, the mortality to incidence ratio is relatively low in comparison to other common cancers. Moreover, the primary treatment modalities, radical prostatectomy and radiation therapy, are associated with a substantial reduction in quality of life. Consequently, there is a growing interest in focal therapy which aims to treat the target tumor while minimizing damage to surrounding tissue.

Focal laser ablation (FLA) is a form of focal therapy in which a laser fiber is inserted into a target tumor and oncologic control is achieved through thermally induced coagulative necrosis. Many groups have performed FLA using magnetic resonance imaging (MRI) for laser fiber targeting and real-time feedback. While this approach has shown promise, we contend that the resource intensive nature of MRI will forever preclude widespread adoption of FLA.

This thesis presents a concerted effort to translate FLA from the MRI suite to the urology clinic. To this end we performed a clinical trial using magnetic resonance – ultrasound fusion guidance and interstitial thermal probes for treatment targeting and monitoring respectively. This approach proved to be safe and potentially effective; however, the utility of thermal probes was found to be inherently limited due to the need for empirically derived thermal damage models.

In an effort to provide an improved monitoring modality we developed an interstitial optical monitoring system that directly assesses the state of tissue based on laser-tissue interaction. To correlate the optical signal and the growth of the coagulation zone we created a tissue mimicking phantom which simulates the thermal and optical response of prostatic tissue and facilitates visualization of the coagulation zone on MRI. FLA was performed in the phantom under MRI surveillance with simultaneous interstitial optical monitoring resulting in the development of a real-time feedback algorithm. The algorithm was subsequently tested in ex vivo bovine tissue and was capable of identifying the coagulation border with a mean absolute error of 0.3�0.1mm. Further work is necessary to demonstrate the utility of interstitial optical monitoring in vivo.