Permanent prostate implants are a type of radiation therapy in which a high dose of radiation is delivered to cancerous tissue by many small radioactive "seeds". Studies of early-stage prostate cancer patients treated in this way and followed for 10 - 12 years indicate a cure rate of about 80%. This result is similar to surgery and other forms of radiotherapy, but comes with fewer side effects and greater convenience for the patient. Further studies show that the radiation dose delivered is the most important factor in achieving a cure. At present this dose is estimated by a computer, using a computed tomography (CT) scan of the patient and a simple calculation method. The dose estimate is not as accurate as it could be, however, because the precise extent of the prostate is hard to determine from the CT scan, and the calculation method does not make use of information about patient body tissues. The researchers propose to eliminate these inaccuracies by using magnetic resonance imaging (MRI) to identify the prostate gland and by developing an improved dose calculation algorithm that includes information about patient tissues. This new approach will allow physicians to assess implant quality with greater certainty, improve their implant technique, and ultimately increase the cure rate to as much as 95%.

This study addresses three major sources of post-implant dosimetry inaccuracy for permanent prostate implants: post-operative edema, prostate contour delineation, and dose calculation method. It is hypothesized that a pragmatic edema model can minimize the first uncertainty, co-registered CT + MR images the second, and an improved dose calculation algorithm the third.

Detailed objectives are to:

• measure and model the effects of edema on dosimetry;
• evaluate CT + MR image registration methods;
• compare dosimetry for CT alone vs. CT + MRI using the contemporary TG-43 dose calculation method;
• set up a Monte Carlo code that makes full use of the information in CT + MR images to perform implant dose calculations;
• compare prostate dosimetry for the Monte Carlo vs. the simpler TG-43 method;
• develop an analytical post-implant dose calculation algorithm for routine clinical use (Monte Carlo is too slow on a single-CPU brachytherapy planning computer); and finally
• assess the performance of the new algorithm.

Of the estimated 250,000 new cases of prostate cancer in North America in 2004, most are early stage disease as a consequence of PSA testing. Permanent prostate implant therapy is a major option for this group, as long-term clinical studies indicate a cure rate equal to surgery and external beam radiotherapy, but with fewer complications. By dealing with dosimetric inaccuracies, a proven treatment can reach its full potential.

Inclusion Criteria:

• Candidate for permanent prostate implant

Exclusion Criteria:

• Not ambulatory