The effect of patient position and treatment technique in conformal treatment of prostate cancer

Purpose: The relative value of prone versus supine positioning and axial versus nonaxial beam arrangements in the treatment of prostate cancer remains controversial. Two critical issues in comparing techniques are: 1) dose to critical normal tissues, and 2) prostate stabilization. Methods and Materials: Ten patients underwent pretreatment CT scans in one supine and two prone positions (flat and angled), To evaluate normal tissue exposure, prostate/seminal vesicle volumes or prostate volumes were expanded 8 mm and covered by the 95% isodose surface by both 6-field axial and 4-field nonaxial techniques. A total of 280 dose-volume histograms (DVHs) were analyzed to evaluate dose to rectal wall and bladder relative to patient position and beam arrangement. A CT scan was repeated in each patient after 5 weeks of treatment. Prostate motion was assessed by comparing early to late scans by three methods: 1) center of mass shift, 2) superior pubic symphysis to anterior prostate distance, and 3) deviation of the posterior surface of the prostate. Results: For prostate (P) or prostate/seminal vesicle (P/SV) treatments, prone flat was;advantageous or equivalent to other positions with regard to rectal sparing. The mechanism of rectal sparing in the prone position may be related to a paradoxical retraction of the rectum against the sacrum, away from the P/SV, Although there was no clear overall preference for beam arrangement, substantial improvements in rectal sparing could be realized for individual patients. In this limited number of patients, there was no convincing evidence prostate position was stabilized by prone relative to supine position. Conclusions: Prone flat positioning was advantageous over other positions and beam arrangements in rectal sparing. This study suggests that patient position is a more critical a factor in conformal therapy than beam arrangement, and may improve the safety of dose escalation. (C) 1999 Elsevier Science Inc.