Inr .I Radmwn Oncoloy,~ Bwl Phv Vol. 20. pp. I3 I7- I324 0360-3016191 $3.M) t .oO Printed ,n the U.S.A All rights reserved Copyright D I99 I Pergamon Press plc ?? Technical Innovations and Notes zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFE TREATMENT PLANNING ISSUES RELATED TO PROSTATE MOVEMENT IN RESPONSE TO DIFFERENTIAL FILLING OF THE RECTUM AND BLADDER R. K. TEN HAKEN, PH.D., J. D. FORMAN, M.D., D. K. HEIMBURGER, M.D., A. GERHARDSSON, M.S., D. L. MCSHAN, PH.D., C. PEREZ-TAMAYO, M.D., S. L. SCHOEPPEL, M.D. AND A. S. LICHTER, M.D. Radiation Oncology Department, University of Michigan, Ann Arbor, MI Conventional simulation for patients with localized prostatic carcinoma often includes opacification of the dose limiting adjacent normal tissues. However, CT-based treatment planning is performed with the bladder and the rectum naturally filled or emptied. These latter conditions more closely approximate those in place at treatment. Comparison of these CT-based treatment plans to simulator films taken with the rectum and bladder opacified yielded indirect evidence of movement of the prostate gland by 0.5 cm or more in 31 of 50 consecutive patients. The range of motion was 0 to 2 cm with an average of 0.5 cm (1.0 cm in the 31 patients). Six additional patients (five with local recurrence following I-125 seed implantation) were analyzed separately using CT scans. Registered CT images (3 mm slices) taken with the rectum and bladder full and/or empty provided direct evidence of prostate movement in 3 of the 6 patients. The dosimetric consequences of this movement are demonstrated using 3-dimensional dose distributions. Prostate movement, Computerized 3-D treatment planning, Beam’s-eye-view (BEV) planning, Conformal therapy. INTRODUCTION The prostate is an anatomic site often considered for ra- diation therapy using arrangements of beams designed to conform a dose distribution closely to a target volume (conformal therapy). The anatomy in the treatment region is well defined. The target volume includes the prostate gland and periprostatic tissues. The dose limiting normal tissues (rectum and bladder) are likewise easily defined. Treatment techniques ( 1-3, 11, 16, 17, 19) have been designed on the assumption that these structures are sta- tionary. The goal of conformal therapy is to individualize treat- ments through use of multiple fields, each having a shape which conforms to the target volume with as little margin as possible. An analysis of the potential benefits of con- formal therapy must coincide with a careful analysis of the magnitude of errors and uncertainties associated with these techniques. Any potential benefit of tight dose mar- gins may be obscured if the dose is not localized around the target because of a change in its position (7, 10, 18). Motion of organs within the target volume and patient set-up uncertainties and movements (not discussed here) now become critical. The University of Michigan approach to the treatment of prostate cancer (22) relies on CT scans obtained with the rectum in a normal or empty state and the bladder opacified with only a minimal amount of IV contrast. Traditionally, patients are simulated with the rectum and bladder partially filled with radio opaque materials (3). Furthermore, patients are treated on a daily basis with variable filling of the rectum and bladder. Any expansion of the rectum or bladder has the potential for moving the prostate plus seminal vesicle target volume. There are, to our knowledge, no reports in the literature describing this motion. This study investigated the degree of movement of the prostate as a function of bladder and rectum filling, as well as the dosimetric consequences of that motion. METHODS AND MATERIALS The University of Michigan’s CT-based approach to designing treatment for carcinoma in the prostate has been Presented in part at the American Society of Therapeutic Ra- diology and Oncology, October l-5, 1989, San Francisco, CA. Reprint requests to: Randall K. Ten Haken, Ph.D., University of Michigan Medical Center, Department of Radiation Oncol- ogy, 1500 E. Medical Center Dr., Ann Arbor, MI 48 109-0010. Acknowledgements-The authors wish to thank Mr. Mark LaVigne and Ms. Rebecca Tesser for their expert assistance in the treatment planning phases of this study. Accepted for publication 14 December 1990. zyxwvutsrqponmlkj 1317