CLINICAL INVESTIGATION Prostate A CONE BEAM CT-BASED STUDY FOR CLINICALTARGET DEFINITION USING PELVIC ANATOMY DURING POSTPROSTATECTOMY RADIOTHERAPY TIMOTHY N. SHOWALTER, M.D., A. OMER NAWAZ, M.A., YING XIAO,PH.D., JAMES M. GALVIN, D.SC., AND RICHARD K. VALICENTI, M.D. Department of Radiation Oncology, Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA Purpose: There are no accepted guidelines for target volume definition for online image-guided radiation therapy (IGRT) after radical prostatectomy (RP). This study used cone beam CT (CBCT) imaging to generate information for use in post-RP IGRT. Methods and Materials: The pelvic anatomy of 10 prostate cancer patients undergoing post-RP radiation therapy (RT) to 68.4 Gy was studied using CBCT images obtained immediately before treatment. Contoured bladder and rectal volumes on CBCT images were compared with planning CT (CT ref ) volumes from seminal vesicle stump (SVS) to bladder–urethral junction. This region was chosen to approximate the prostatic fossa (PF) during a course of post-RP RT. Anterior and posterior planning target volume margins were calculated using ICRU report 71 guidelines, accounting for systematic and random error based on bladder and rectal motion, respectively. Results: A total of 176 CBCT study sets obtained 2 to 5 times weekly were analyzed. The rectal and bladder borders were reliably identified in 166 of 176 (94%) of CBCT images. Relative to CT ref , mean posterior bladder wall position was anterior by 0.1 to 1.5 mm, and mean anterior rectum wall position was posterior by 1.6 to 2.7 mm. Calculated anterior margin as derived from bladder motion ranged from 5.9 to 7.1 mm. Calculated posterior margin as derived from rectal motion ranged from 8.6 to 10.2 mm. Conclusions: Normal tissue anatomy was definable by CBCT imaging throughout the course of post-RP RT, and the interfraction anteroposterior motion of the bladder and rectum was studied. This information should be considered in devising post-RP RT techniques using image guidance. Ó 2008 Elsevier Inc. Cone beam, Postprostatectomy, Prostate cancer, Radiation therapy, Image guidance. INTRODUCTION Radiation therapy (RT) is delivered after radical prostatec- tomy (RP) either as salvage treatment for an elevated pros- tate-specific antigen (PSA) level (1–6) or as adjuvant therapy for patients with high-risk pathologic features (7, 8). Recent prospective data demonstrated a disease-free sur- vival benefit of adjuvant RT for pathologic T3N0 prostate cancer (9, 10). Despite literature supporting the delivery of post-RP RT to the prostatic fossa (PF), no clear target defini- tion guidelines exist for intensity-modulated radiation ther- apy (IMRT) or image-guided RT (IGRT) (11). Visualization of the PF is limited on standard CT images, with significant interobserver variability and uncertainty in clinical target volume (CTV) definition (12). Efforts to incor- porate complementary imaging modalities such as MRI for PF target volume definition have generated neither demon- strably more reliable PF delineation nor practical contouring guidelines (13). Regardless of the imaging modality, direct visualization and delineation of the PF CTV is fraught with uncertainty. On the other hand, it is possible to distinguish the borders of important nearby pelvic structures—namely, the bladder and the rectum. The reliability of rectal volume definition on helical CT is supported by analysis of rectal contours defined in a prospective trial, suggesting the feasi- bility of rectal dose-volume data collection in a multicenter setting (14). Fiorino et al. (15) described a correlation be- tween PF CTV shift and anterior rectal wall shift for the cra- nial half of the rectum in their report of rectal and bladder movement during post-RP RT using weekly CT images. These studies support the reliability of CT-defined rectal con- tours and a limited correlation between PF CTV and anterior rectal wall, an important tenet in the current study. The data reported by Fiorino et al. are limited by the infre- quency of image collection and the acquisition of images at a time and place separate from the treatment couch. Although Reprint requests to: Richard K. Valicenti, M.D., Department of Radiation Oncology, Thomas Jefferson University Hospital, 111 S. 11th Street, Philadelphia, PA 19107; Tel: (215) 955-6702; Fax: (215) 955-0412; E-mail: richard.valicenti@mail.tju.edu Conflict of interest: none. Acknowledgment—The study was presented to the 48th Annual Meeting of the American Society for Therapeutic Radiology and Oncology (ASTRO), November 5–9, 2006, Philadelphia, PA. Received March 8, 2007, and in revised form June 14, 2007. Accepted for publication June 15, 2007. 431 Int. J. Radiation Oncology Biol. Phys., Vol. 70, No. 2, pp. 431–436, 2008 Copyright Ó 2008 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/08/$–see front matter doi:10.1016/j.ijrobp.2007.06.026