doi:10.1016/j.ijrobp.2005.12.033 PHYSICS CONTRIBUTION INFLUENCE OF INTRAFRACTION MOTION ON MARGINS FOR PROSTATE RADIOTHERAPY DALE W. LITZENBERG, PH.D.,* JAMES M. BALTER, PH.D.,* SCOTT W. HADLEY, PH.D.,* HOWARD M. SANDLER, M.D.,* TWYLA R. WILLOUGHBY, M.S.,��� PATRICK A. KUPELIAN, M.D.,��� AND LISA LEVINE, PH.D.��� *Department of Radiation Oncology, University of Michigan, Ann Arbor, MI ���M. D. Anderson Cancer Center Orlando, Orlando, FL and ���Calypso Medical Technologies, Seattle, WA Purpose: To assess the impact of intrafraction intervention on margins for prostate radiotherapy. Methods and Materials: Eleven supine prostate patients with three implanted transponders were studied. The relative transponder positions were monitored for 8 min and combined with previously measured data on prostate position relative to skin marks. Margins were determined for situations of (1) skin-based positioning, and (2) pretreatment transponder positioning. Intratreatment intervention was simulated assuming conditions of (1) continuous tracking, and (2) a 3-mm threshold for position correction. Results: For skin-based setup without and with inclusion of intrafraction motion, prostate treatments would have required average margins of 8.0, 7.3, and 10.0 mm and 8.2, 10.2, and 12.5 mm, about the left���right, anterior��� posterior, and cranial���caudal directions, respectively. Positioning by prostate markers at the start of the treatment fraction reduced these values to 1.8, 5.8, and 7.1 mm, respectively. Interbeam adjustment further reduced margins to an average of 1.4, 2.3, and 1.8 mm. Intrabeam adjustment yielded margins of 1.3, 1.5, and 1.5 mm, respectively. Conclusion: Significant reductions in margins might be achieved by repositioning the patient before each beam, either radiographically or electromagnetically. However, 2 of the 11 patients would have benefited from continuous target tracking and threshold-based intervention. �� 2006 Elsevier Inc. Prostate cancer, Intrafraction motion, Margins, Setup correction, Organ motion. INTRODUCTION The goal of three-dimensional conformal therapy is to shape the dose distribution to the prescribed target volume as closely as possible without sacrificing target coverage. This technique results in the sparing of neighboring healthy tissues and often leads to fewer complications and higher quality of life. At the same time, this technique might allow higher doses to target volumes that are limited by toxicity of normal tissues, potentially resulting in better local tumor control. To ac- count for daily setup error and internal motion of the clinical target volume (CTV), an additional margin is added to form the planning target volume (PTV). Historically, site-specific PTV margins have been derived from studies of populations of patients, according to multiple CT scans, daily imaging, and clinical experience. In one study, population-based mar- gins were developed to ensure that 90% of the patients must receive a minimum of 95% of the prescribed dose within the PTV (1). By this rule, population margins will allow some patients to be under-dosed, and other patients will have larger margins than necessary. In the past decade, many researchers have explored meth- ods to customize treatment margins to the individual pa- tient, to maximize the benefits of conformal therapy. Two general approaches have emerged. Online approaches seek to localize the target volume each day by imaging hard or soft tissues or implanted fiducial markers (2���7). Offline approaches seek to determine and correct the systematic errors of each patient���s target volume position while adap- tive therapy also accounts for that individual���s random setup error (8���13). The daily online localization approach offers the possibility of minimizing systematic and random target volume positioning errors for every patient, at the expense Reprint requests to: Dale Litzenberg, Ph.D., University of Mich- igan Health System, Department of Radiation Oncology, UH - B2C432C Box 0010, 1500 East Medical Center Drive, Ann Arbor, MI 48109-0010. Tel: (734) 936-1934 Fax: (734) 936-2261 E-mail: litzen@umich.edu Presented at the 47th Annual Meeting of the American Society for Therapeutic Radiology and Oncology (ASTRO), October 16��� 20, 2005, Denver, Colorado. This work was supported by grant P01-CA59827 from the National Institutes of Health. J.M.B. is a scientific consultant to Calypso Medical Technolo- gies. L.L.L. is an employee of Calypso Medical Technologies. Both have a financial interest in Calypso Medical Technologies. Received Aug 8, 2005, and in revised form Oct 14, 2005. Accepted for publication Dec 5, 2005. Int. J. Radiation Oncology Biol. Phys., Vol. 65, No. 2, pp. 548���553, 2006 Copyright �� 2006 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/06/$���see front matter 548

of requiring a large amount of time and effort. The offline and adaptive approaches require a modest amount of time and effort in the first week or two of treatment to gather enough data to make reasonable estimates of each individual patient���s systematic offsets and random variations, which are then used to customize the remainder of the treatment. Although online and offline approaches offer the possibility of realizing further gains from conformal radiotherapy, they tend to neglect target volume motion that might occur after pretreatment setup, while therapy is delivered. In other studies, investigators have sought to minimize this intrafraction motion by using favorable patient posi- tioning or target volume immobilization devices in combi- nation with pretreatment target volume localization (3, 14��� 21). Only in the past several years have fluoroscopic X-ray systems been investigated for real-time monitoring of in- trafraction target volume motion (22���30). In the work pre- sented here, the PTV margins required to account for in- trafraction target volume motion, under various setup and monitoring conditions, were studied according to data from real-time (approximately 10 Hz) electromagnetic tracking with the Calypso 4D Localization System (31, 32) (���Ca- lypso System��� Calypso Medical, Seattle, WA). METHODS AND MATERIALS Data acquisition Historically, PTV setup has been accomplished by aligning skin marks to room-based lasers or by using portal imaging to align bony anatomy to digitally reconstructed radiographs calculated from the planning CT. However, in many institutions, implanted gold markers have become a standard of care. The location of these markers can be determined daily through portal imaging, to deter- mine the pretreatment preparation and treatment errors. By using the skin-mark setup technique, immediately followed by pretreat- ment localization of implanted markers, the combined interfraction patient setup and internal target volume motion error can be determined. These data were used to retrospectively determine the appropriate treatment margins under the assumption of no in- trafraction target volume motion. The effects of intrafraction motion on the appropriate margins were estimated according to prostate motion data obtained during an internal review board���approved pilot study involving the in- vestigational Calypso System (32). These data were combined with the previously measured data on prostate position relative to skin marks to determine the appropriate margins when setting up to skin marks, once before treatment, both (1) without and (2) with intrafraction motion. When just the measured intrafraction motion data were used, margins were determined (3) when setting up to implanted markers, once before treatment, (4) when setting up to implanted markers, before delivery of each treatment beam, and (5) when setting up to markers, before delivery of each treatment beam, followed by continuous marker tracking (10 Hz) with a 3-mm threshold for beam cutoff, repositioning, and resumption of treatment. The above data acquisition and analysis are described in detail below. Positioning of prostate target volume by skin marks In general, all prostate patients undergoing radiotherapy are considered for gold marker insertion before simulation and treat- ment. Those with relative contraindications to the procedure are excluded (i.e., those using anticoagulants). Initially, the use of gold markers was performed on a clinical trial. Currently, the procedure is a routine part of treatment planning and delivery, and approxi- mately 25% of patients have gold markers inserted. Three 18-carat gold markers (0.9 mm 5 mm) were implanted in the prostate transrectally under ultrasound guidance. Patients were supine on a flat pad with a support under their knees and their feet fixed together for their planning CT and treatments. After treatment planning, the three-dimensional coordinates of each marker rela- tive to isocenter were stored in a file for daily reference during pretreatment setup. Patients were given no rectal or bladder filling instructions. Each day, the patient���s skin tattoos were aligned to the isocenter lasers. Sequential orthogonal megavoltage digital radiographs were obtained with the Varian portal imager (Varian Medical Systems, Palo Alto, CA [model AS500 IAS2]). Setup errors in the left���right (LR) and anterior���posterior (AP) directions were obtained through Portal Vision (Varian Medical Systems) template alignment of the anterior and lateral radiographs, respec- tively, whereas the inferior���superior (IS) setup errors were an average of the results from both images. The setup errors of the geometrical centroid of the markers are summarized in Table 1. Measurement of intrafraction prostate motion The Calypso System is an investigational device that allows continuous electromagnetic tracking of implanted markers (Bea- con transponders [Calypso Medical, Seattle, WA]). Data collected from initial trials using this system were studied to assess the relative value for intrafraction intervention during prostate radio- therapy. Under internal review board��� approved protocols, 20 patients underwent implantation of three transponders in the pros- tate. Eleven of these patients were eligible and consented for participation in additional monitoring and localization and pro- vided evaluable localization and tracking data. The transponders are 8 mm long by 1.85 mm in diameter (compared with gold markers, which can be up to 5 mm long and 1 mm in diameter) and are implanted transrectally with a 14-gauge needle, by procedures similar to those used for obtaining prostate biopsies. Transponders were placed at the apex and right and left base under ultrasound guidance. An isocenter was chosen relative to the geometric center of the transponders on CT, and the (supine) patients were posi- tioned to this isocenter with the electromagnetic system. After positioning (and a 3���5-min delay to permit radiographic position verification), the relative positions of the implanted transponders were monitored at 10 Hz for approximately 8 min, which was estimated to be the length of time required to deliver a fraction of Table 1. Average, standard deviation of the average, and standard deviation of marker setup error in each direction, when each supine patient was set up to skin tattoos using isocenter lasers Skin-to-marker setup error (cm) LR AP IS Mean 0.013 0.063 0.005 s,Inter 0.223 0.146 0.305 s,Inter 0.344 0.523 0.333 Abbreviations: LR left���right AP anterior���posterior IS inferior���superior. 549 Influence of intrafraction motion on prostate margins ��� D. W. LITZENBERG et al.