Implementation of daily image-guided radiation therapy using an in-room CT scanner for prostate cancer isocentre localization

Transcription

1 Journal of Medical Imaging and Radiation Oncology 53 (2009) RADIATION ONCOLOGY TECHNICAL ARTICLE Implementation of daily image-guided radiation therapy using an in-room CT scanner for prostate cancer isocentre localization K Knight, N Touma, L Zhu, GM Duchesne and J Cox Tattersall s Cancer Centre, Richmond, Victoria, Australia K Knight HScD; N Touma BAppSc; LZhu PhD; GM Duchesne MBChB, FRCR (London), MD, FRANZCR; JCoxPhD. Correspondence Dr Kellie Knight, Radiation Therapy Services, Peter MacCallum Cancer Centre, Locked Bag 1, A Beckett Street, Melbourne, Vic. 8006, Australia. kellie.knight@petermac.org Conflicts of interest: None. Submitted 8 June 2008; accepted 1 August doi: /j x Summary The Tattersall s Cancer Centre has been performing image-guided radiation therapy (IGRT) using an in-room CT on rails since 2003 to verify accurate patient setup position (relative to bony anatomy) immediately prior to treatment delivery for prostate cancer patients. While the concept of online correction for bony anatomy is well established, the use of an in-room CT scanner also enables the collection and offline analysis of soft tissue volumetric data. Although initially IGRT was implemented under a research protocol, in-room CT verification has continued to be used to measure and correct for patient setup variations for all patients undergoing intensity modulated radiation therapy (IMRT) treatments. The present paper outlines the protocol that was used to implement IGRT using an in-room CT scanner at the Tattersall s Cancer Centre. Online corrections that minimize patient setup uncertainties allow confidence in delivering dose escalation as well as decreasing the margins required around the target volume. With improvements in auto-contouring tools, IGRT will also have the ability to measure and correct for variations in target and critical structure positioning online, rather than the current offline methods utilized. Key words: computed tomography; image-guided radiation therapy; in-room CT scanner; prostate cancer; radiotherapy treatment verification; setup verification. Introduction In Australia, external beam radiation therapy treatment protocols for prostate cancer typically deliver up to 78 Gy using 3-D conformal or intensity modulated radiation therapy (IMRT) plans. During this seven- to eight-week course of radical treatment several variables can affect the accuracy with which the patient s treatment can be delivered. These include daily patient setup uncertainties, organ volume uncertainties and organ motion. To ensure that treatment is accurately delivered to the planned volume, regular treatment verification is required. The Faculty of Radiation Oncology Genito-urinary Group recommended that an isocentre check using AP and Lateral films is acquired at least weekly during treatment, and ideally daily during the first week of treatment. If available, daily localization with fiducial markers or ultrasound/ct imaging is preferred. 1 In line with these recommendations it has been reported that in early 2003 the majority of centres in Australia and New Zealand performed verification for prostate cancer patients using weekly orthogonal films, although the results of this survey indicated that verification protocols incorporating more intensive imaging during the first week of treatment, with or without additional follow-up, were being introducing into clinical practice. 2 As patient setup accuracy is measured relative to bony landmarks, orthogonal port films, electronic portal images (EPI) or repeat computed tomography (CT) scans can be used to assess setup uncertainty. The translation required to align the pelvic bones between planning and treatment images can be accurately calculated by software programs or performed manually by utilizing measuring tools available on the imaging program. Sequential images can be used to break down setup uncertainty into random and systematic components and adjust the patient s setup accordingly. This information can then be used to calculate the clinical tumour volume (CTV) to planning target volume (PTV) margins required in each clinical practice. However, the benefit 132

2 Implementing IGRT into clinical practice of using an in-room CT scanner is the ability to move away from standard PTV margins. In-room CT enables the introduction of adaptive radiotherapy where the treatment fields can be optimized by accounting for interfraction changes in the position and shape of the target. 3 It also enables easier identification and correction of rotational setup variation when combined with a sixdegrees-of-freedom robotic couch. While orthogonal port films or EPI have the ability to detect patient setup variations, in 2003 no centres in Australia and New Zealand reported using imaged guidance to assess the other sources of uncertainty present throughout a course of radiation therapy. 2 The commercial availability of online imaging techniques, such as inroom kilovoltage CT scanners on rails, transabdominal ultrasound, kilovoltage or megavoltage cone beam CT (CBCT) or even using orthogonal films to track implanted markers has recently led to an increase in research studies utilizing IGRT. Although radio-opaque fiducial markers inserted into the prostate have become popular and have overcome the limitations of orthogonal imaging by tracking prostate motion, the insertion of fiducial markers into the prostate requires a highly invasive procedure and has limited applicability to sites outside the pelvis. 4 Therefore, in order to assess internal organ volume uncertainties, organ motion and their effect on the dose delivered CT imaging is the preferred option. A review of the published work found that prior to the availability of in-room CT scanners, image-guided radiation therapy (IGRT) research using repeat CT scanning for prostate cancer was performed over a limited number of fractions and imaged only a small number of patients due to the time and inconvenience of repeating the scans in the planning department. Of the 24 articles published on the use of out-of-treatment room CT scans, only six utilized weekly CT images, with a maximum of 19 patients included in one study Weekly CT scans allow observation of the patient in 15 25% of treatment sessions, thus providing only a representative sample of the range and frequency of volume and motion variations. In 2005 Wong et al. published the only known article on daily in-room CT verification during the last week of treatment for prostate cancer patients; however, to date, no published studies have assessed these uncertainties on a daily basis throughout an entire treatment course. 29 In 2003 the Tattersall s Cancer Centre commenced a clinical trial on the impact of daily CT-guided verification for prostate cancer. The study utilized CT images to perform daily online corrections relative to bony anatomy for 25 patients; however, all images were assessed offline to determine the potential benefits of implementing online corrections using 3-D soft tissue volumetric data into clinical practice. The IGRT study conducted at the Tattersall s Cancer Centre has allowed a more comprehensive assessment of variables that effect treatment accuracy on a larger prostate cancer patient population than most published studies. However, the full results of the present study are not presented here. Instead, the aim of the present paper is to document the protocol used to implement daily IGRT using an in-room CT scanner at the Tattersall s Cancer Centre and assess the feasibility of this protocol for patients receiving radical treatment for prostate cancer. Methods When the Tattersall s Cancer Centre (operated in partnership with the Peter MacCallum Cancer Centre) opened in 2003 it was the first centre in the Southern Hemisphere to clinically implement a linear accelerator with in-room CT capabilities. The Siemens Primatom (Siemens Medical Systems, Germany), consisting of a Primus linear accelerator (Siemens Medical Systems, Germany) and a Somatom Balance CT scanner (Siemens Medical Systems, Germany) mounted on rails, enabled online IGRT to be implemented into clinical practice by scanning the patient in the treatment position immediately prior to treatment delivery. At the time of purchase in 2002, the Somatom Balance CT scanner cost an additional $A (at an exchange rate of $A1= $US0.51) to convert the Primus to a Primatom. It took an extra week to commission the machine, including verification of the geometry, and requires an additional 30 min per day quality assurance (QA) time to maintain the CT scanner. Additional data storage was not explicitly purchased because there was sufficient space on the CT hardware, although backup tapes have been used to transfer data no longer required. A standalone CT scanner linac combination, such as the Primatom, integrates existing imaging technology and therapy. The standalone approach, as opposed to cone beam technology, offers diagnostic quality volumetric imaging while maintaining the reference between verification and treatment by using a pivoting patient Fig. 1. Primatom linear accelerator. Linac combined with CT on rails. 133

3 K Knight et al. couch. The linac and the CT gantry are positioned at opposite ends of the couch so that by rotating the treatment couch, treatment or CT scanning can be performed (Fig. 1). The rotational axis of the linac gantry is coaxial with that of the CT gantry, and the position of the linac isocentre on the couch should match the origin of the coordinate system for the CT scanning when the couch is rotated 180 toward the CT side. Twenty five patients who were prescribed radical external beam radiation therapy for treatment of prostate cancer were enrolled into a CT verification study between October 2003 and October The study was approved by the Peter MacCallum Cancer Centre, Epworth Hospital and University of Sydney ethics committees. The aim of the study was to comprehensively assess the uncertainties associated with prostate cancer radiotherapy (such as bladder and rectal volume reproducibility, internal organ motion, setup error and their effect on the dose delivered) to determine the potential benefit of implementing online corrections for soft tissue anatomy. For each of the 25 patients enrolled in the study, a daily CT of the entire pelvis was acquired prior to treatment. Therefore, prior to the commencement of this research project, a protocol was required for the use of the in-room CT scanner that was quick and efficient for the radiation therapists (RT) to implement on a daily basis. The method developed is described in the results section below. Results The method developed to enable daily online isocentre correction relative to bony anatomy is demonstrated in Figure 2 and is described in greater detail below. Patient set-up to external marks using lasers at linac Rotate couch to CT side and acquire topogram Determine iso position sup/inf, take single slice CT image through central plane Reposition patient according to bony anatomy Acquire 3D CT volumetric dataset of pelvis Rotate couch back to linac position verify patient position with lasers Treatment delivered Fig. 2. Schematic diagram of the in-room CT verification process. Documentation of the in-room CT verification protocol At the start of each treatment fraction the RT positioned, localized and marked up each patient on the treatment couch according to the standard treatment protocol. However, before treatment was delivered, the isocentre location relative to bony landmarks was verified using the in-room CT. To facilitate this, ball bearings (BB) were placed on the patient s skin at the laser position while the couch was located under the linear accelerator. The digital couch readout was zeroed at this planned isocentre position to give a relative digital couch readout. The patient couch was then rotated to 180.1, asdisplayed on the digital couch readout, in preparation for the pre-treatment CT scan. Although, in theory, a couch rotation of exactly 180 was required, during acceptance testing it was determined that for the CT to travel parallel over the couch it was necessary to rotate to The Primatom CT gantry moves along rails for scanning rather than moving the table as in conventional CT. This is designed to eliminate patient movement during the scan and to minimize errors caused by flexures in the treatment couch. The CT on rails design decreases the risk of errors due to mechanical movement because the only movement during the CT scan is the rotation of the couch. Once adequate clearance of the CT over the patient was confirmed by the RT, the CT was zeroed at the central axis so that the CT lasers intersected the BB placed on the patient s skin while under the linac. Using the BB placed on the patient s skin at the isocentre, the CT was reviewed for positional accuracy using a topogram and control slice (Fig. 3). The topogram was used to assess the location of the isocentre in the superior inferior direction. Once the isocentre located in this direction was confirmed, a control slice through the plane of the isocentre was used to assess the position of the isocentre in the anterior posterior and lateral directions. This process was simplified by the fact that prostate cancer patients at all Peter MacCallum Cancer Centre sites have their isocentre located at a standardized location rather than at the centre of the target volume. Since the isocentre location was always set 1.0 cm inferior to the upper border pubis (UBP), in the midline and at a depth of 6 cm posterior to the anterior edge of the pubis bone, the potential for errors in determining isocentre displacements was decreased. The BB placed on the patient s skin were easily identifiable on the CT scans and were used to verify isocentre location rather than the CT origin. This was because during the couch rotation the couch height dropped 0.45 cm due to the rotation mechanism of the couch. This was investigated prior to the commencement of the study and it was confirmed that there was no associated change in couch tilt. However, the change in couch height meant that the CT lasers in the coronal plane did not intersect the BB placed on the skin when the couch was 134

4 Implementing IGRT into clinical practice Fig. 3. Topogram (left) and control scan (right) were used to assess isocentre accuracy in 3-D. The crosshair function used to measure the isocentre location relative to the pubic bone is shown on the control scan. under the linear accelerator. Therefore, if the origin of the CT scan was used to confirm the isocentre setup, a correction of 0.45 cm would have been required, thus introducing the potential for calculation error by the treatment RT. At the CT console the crosshair function was used to align the BB placed on the patient s skin. Using the measuring tool (ruler) on the CT console, the position of the isocentre in relation to the reference landmarks was measured. Variations of <0.5 cm required no pretreatment correction; however, the coordinates of the isocentre for each fraction were recorded so that patient setup accuracy could be determined. If the isocentre placement was outside the accepted tolerance (0.5cm) an online isocentre correction was applied. The RT used the relative digital couch readout inside the treatment room to move the required distance. The BB were then adjusted on the patient s skin to reflect the change in isocentre position and all corrections were recorded. Once the isocentre position for treatment was established, a helical CT data set was obtained from +12 cm to 10 cm, using a 5-mm slice width and a pitch of 1.5 (130 kv, 180 mas). The couch was then rotated back to 0 and the radiation treatment was delivered. The added benefit of using the BB as external markers on the skin was that when the patient was rotated back under the linac from the CT, it provided increased confidence in the accurate location of the isocentre for the RT by allowing a visual check that the verified isocentre location (as represented by the BB) matched the linear accelerator laser system. Once treatment was completed the axial CT images were exported back to the CMS XiO (Computerised Medical Systems, Maryland Heights, MO, USA) planning computer system for offline assessment. Thecompleteresultsofthisresearch,includinginternal organ volume variations, organ motion, and dose reproducibility, are not presented in the present paper. Feasibility of in-room CT verification During the course of the present study, 816 online CT verification scans on 25 patients were performed allowing IGRT to be implemented at the Tattersall s Cancer Centre. An important outcome of the study was to assess the time and resource requirements to determine whether the use of the in-room CT scanner was feasible. When IGRT was initially implemented patients were allocated 30-min appointments for each fraction; however, as the staff became familiar with the process, the time taken to complete each treatment was significantly shorter. By the end of the study patients were allocated only min per IGRT session. A comparison of the time taken to deliver a standard treatment with either CT or EPI verification was completed to assess whether the in-room CT protocol was efficient compared to the standard EPI verification imaging protocols. The time taken to complete each verification method was recorded once the patient was correctly positioned on the treatment couch. Table 1 demonstrates the steps for both verification procedures included in the time comparison. When taking into account the time taken by two RT to review, assess and action any changes from the EPI in an offline environment, the average time taken using EPI (14 min) was longer than that needed to complete online IGRT using the in-room CT scanner (11.5 min). When an online correction was performed using the in-room CT, the total time taken for each patient was at the upper end of the range shown in Table 1 ( min). However, throughout the entire study only 10% of treatments required online correction and, therefore, although some patient treatments using in-room CT will take longer than EPI, the impact of this time on resources over an entire course would be minimal. Anecdotal evidence from the staff who treated these patients supports 135

5 K Knight et al. Table 1. Comparison of time taken to complete CT and electronic portal images (EPI) verification procedures CT verification procedure Rotation of the couch through 180 Topogram to assess superior inferior location of the isocentre Control scan to assess anterior posterior and lateral location of the isocentre Isocentre corrections performed (if required) and verified Couch rotated back under the linac Treatment delivery Mean: 11.5 min Min: 8 min; max: 16.5 min Online corrections: 82/816 treatments (10%) max: 12; min: 0; mean: 3.3 EPI verification procedure Two orthogonal EPI acquired (including the time taken for the RT to enter the room and position the imager) No online corrections performed Treatment delivery Offline assessment by two RT to assess trends (e.g. NAL protocols) Mean: 14 min Min: 11 min; max: 17.5 min RT, radiation therapist. the figures presented here. Staff frequently commented on how quick and easy the CT verification process is compared to EPI. This might also be influenced by the fact that CT verification is completed prior to the patient s treatment, thus eliminating the need to re-enter the room once treatment has commenced to position the EPI during the patient s treatment and improves treatment unit workflow as the CT images do not need to be reviewed later in the day. Additional radiation dose As part of the ethics approval process the extra radiation dose delivered to the patients enrolled in the study was considered and measured. Patients underwent up to 37 additional CT scans throughout their treatment course. Each CT scan included a topogram and a single slice to accurately determine the location of the isocentre. Once the isocentre was confirmed, a full CT dataset was collected throughout the pelvis for the purpose of the main research. The Physical Sciences department estimated the approximate additional dose received by the patient as 1.2 Gy over 37 fractions. Although this dose is not negligible, the potential benefit was deemed to justify any risks to the patient. Discussion Image-guided radiation therapy using an in-room CT scanner was easily implemented at the Tattersall s Cancer Centre by simply replicating the processes that were used during the treatment planning stages. The same processes that were used to identify the isocentre location relative to the standard bony landmarks during the planning CT were used to implement IGRT on the treatment machine. The fact that the planning CT scanner and the inroom CT scanner, along with the tools available on both CT consoles, had identical capabilities assisted in the education of, and smooth transition for, the RT who regularly rotated between both treatment and the planning section over the period of the study. Although this IGRT process was initially adopted for prostate patients, it is now also used for other anatomical sites where increased accuracy is required. Although IGRT using in-room CT verification could theoretically correct for internal organ motion online, when the Primatom was installed at the Tattersall s Cancer Centre this potential was limited by the lack of commercially available software to allow corrections to be performed at the CT console. Although this function could be performed on the treatment planning computer system, exporting the images, contouring the required structures, calculating shift magnitudes and implementing any resultant corrections was deemed to be inefficient. However, this analysis was performed in the offline environment so that the impact of IGRT on clinical practice once these tools were available could be established. One concern that IGRT has raised is the increased dose delivered to the patient and how it weighs against the potential benefits that IGRT offers. In our study the increased dose delivered to the patient was quantified prior to implementation and it was decided that the clinical benefit of IGRT was justifiable. Remeijer et al. 30 also compared the dose delivered by CT and EPI, and found that CT correction was shown to actually reduce patient dose, although EPI irradiates a smaller volume. The protocol documented in this article has continued to be implemented for all IMRT patients at the Tattersall s Cancer Centre, although only the topogram and the single slice are required to determine the isocentre placement. This has reduced the dose delivered to the patient. The time, cost and effect on patient waiting lists will ultimately limit the widespread application of daily CT verification and IGRT. Although we have established that the CT verification method currently implemented is more efficient than EPI, the present study has been unable to establish whether online matching for soft tissue anatomy will be as efficient, which is the obvious advantage of using a CT-based system and would, therefore, require further investigation prior to implementation. 136

6 Implementing IGRT into clinical practice IGRT is most likely to be used in cases where increased accuracy is required, such as IMRT, or limited to a small number of fractions, for example, only using IGRT during a boost phase instead of an entire course or using it during the first week of treatment to determine individual patient CTV-PTV margins and then adapting the remaining treatments. However, the impact of IGRT on linac treatment space could be offset in the future by facilitating hypo-fractionated or single dose treatments. 31 One disadvantage of not using IGRT on a daily basis was observed during the introduction of the B-Mode aquisition and targeting (BAT) ultrasound system. The introduction of this new verification technology resulted in therapists paying less attention to the initial setup using the skin marks as they relied more directly on the technology for the final positioning. 32 Therefore, it should be reinforced during the implementation of IGRT that the same attention to detail is required when treating all fractions, regardless of online verification techniques. In theory, in-room CT scanners allow online verification and correction for patient setup errors and organ motion. The use of online IGRT will shift the emphasis from patient localization to immobilization. Because many of the uncertainties that are currently being accounted for in the CTV-PTV margin will be minimized, there will be an increased need to guarantee that the patient does not move during treatment. However, in practice, IGRT capabilities cannot be fulfilled without image analysis software to generate the corrections necessary to reposition the patient. The ability to fuse in-room verification CT scans with the planning CT scan using automatic and manual fusion techniques will allow accurate and speedy assessments of the patient s position and enable the vision of online corrections for internal organ motion to be realized. Conclusion We found that using a topogram and single slice was a fast and efficient method to ensure the planned and treatment isocentres for prostate cancer patients were matched according to bony anatomy. This in-room CT verification method allowed fast and accurate decision making. The introduction of in-room CT verification had only a minor impact on the time resources of the department and is suitable for use when increased isocentre accuracy is required. Acknowledgements The authors wish to acknowledge all the radiation therapists, radiation oncologists and physicists at the Tattersall s Cancer Centre for their enthusiasm in introducing IGRT into clinical practice and for assisting in the collection of research data during the daily treatment of their patients. The corresponding author would also like to express her gratitude to the Division of Radiation Oncology and Department of Radiation Therapy Services at the Peter MacCallum Cancer Centre, Melbourne, for supporting this research throughout her appointment as a Research Radiation Therapist. References 1. Skala M, Berry M, Duchesne G et al. Australian and New Zealand three-dimensional conformal radiation therapy consensus guidelines for prostate cancer. Australas Radiol 2004; 48: Knight KA. Patient positioning and treatment instructions used during radiation therapy of the prostate: results of an Australian and New Zealand survey. Radiographer 2005; 52: Court LE, Dong L, Lee AK et al. An automated CT-guided adaptive radiation therapy technique by on-line modification of multileaf collimator leaf positions for prostate cancer. Int J Radiat Oncol Biol Phys 2005; 62: Mackie TR, Kapatoes J, Ruchala K et al. Image guidance for precise conformal radiotherapy. Int J Radiat Oncol Biol Phys 2003; 56: Melian E, Mageras GS, Fuks Z et al. Variation in prostate position quantification and implications for three-dimensional conformal treatment planning. Int J Radiat Oncol Biol Phys 1997; 38: Martinez AA, Yan D, Lockman D et al. Improvement in dose escalation using the process of adaptive radiotherapy combining with three dimensional conformal or intensity modulated beams for prostate cancer. Int J Radiat Oncol Biol Phys 2001; 50: Zelefsky MJ, Crean D, Mageras GS et al. Quantification and predictors of prostate position variability in 50 patients evaluated with multiple CT scans during conformal radiotherapy. Radiother Oncol 1999; 50: Mechalakos JG, Mageras GS, Zelefsky MJ et al. Time trends in organ position and volume in patients receiving prostate three dimensional conformal radiotherapy. Radiother Oncol 2002; 62: Stroom JC, Kroonwijk M, Pasma KL, Koper PCM, van Dieren EB, Heijmen BJM. Detection of internal organ movement in prostate cancer patients using portal images. Med Phys 2000; 27: Miralbell R, Ozsoy O, Pugleisi A et al. Dosimetric implications of changes in patient repositioning and organ motion in conformal radiotherapy for prostate cancer. Radiother Oncol 2003; 66: Dawson LA, Mah K, Franssen E, Morton G. Target position variability throughout prostate radiotherapy. Int J Radiat Oncol Biol Phys 1998; 42: Tinger A, Michalski JM, Cheng A et al. A critical evaluation of the planning target volume for 3-D conformal radiotherapy of prostate cancer. Int J Radiat Oncol Biol Phys 1998; 42: Roeske JC, Forman JD, Mesina CF et al. Evaluation of changes in the size and location of the prostate, seminal vesicles, bladder, and rectum during a course of 137

7 K Knight et al. external beam radiation therapy. Int J Radiat Oncol Biol Phys 1995; 33: Lebesque JV, Bruce AM, Kroes APG, Touw A, Shouman T, van Herk M. Variation in volumes, dose-volume histograms, and estimated normal tissue complication probabilities of rectum and bladder during conformal radiotherapy of T3 prostate cancer. Int J Radiat Oncol Biol Phys 1995; 33: Zellars RC, Roberson PL, Strawderman M et al. Prostate position late in the course of external beam therapy: patterns and predictors. Int J Radiat Oncol Biol Phys 2000; 47: Beard CJ, Kijewski P, Bussiere M et al. Analysis of prostate and seminal vesicle motion: implications for treatment planning. Int J Radiat Oncol Biol Phys 1996; 34: Rudat V, Schraube P, Oetzel D, Zierhut D, Flentje M, Wannenmacher M. Combined error of patient positioning variability and prostate motion uncertainty in 3D conformal radiotherapy of localised prostate cancer. Int J Radiat Oncol Biol Phys 1996; 35: Antolak JA, Rosen II, Childress CH, Zagars GK, Pollack A. Prostate target volume variations during a course of radiotherapy. Int J Radiat Oncol Biol Phys 1998; 42: Lattanzi J, McNeeley S, Pinover W et al. A comparison of daily CT localisation to a daily ultrasound-based system in prostate cancer. Int J Radiat Oncol Biol Phys 1999; 43: Dehnad H, van der Heide UA, van Moorselaar RJA, Hofman P, Lagendijk JJW. Clinical feasibility study for the use of implanted gold seeds in the prostate as reliable positioning markers during megavoltage irradiation. Radiother Oncol 2003; 67: Crook J, Raymond Y, Salhani D, Yang H, Esche B. Prostate motion during standard radiotherapy as assessed by fiducial markers. Radiother Oncol 1995; 37: Bentel GC, Munley MT, Marks LB, Anscher, MS. The effect of pressure from the table top and patient position on pelvic organ location in patients with prostate cancer. Int J Radiat Oncol Biol Phys 2000; 47: McLaughlin PW, Wygoda A, Sahijdak W et al. The effect of patient position and treatment technique in conformal treatment of prostate cancer. Int J Radiat Oncol Biol Phys 1999; 45: Weber DC, Nouet P, Rouzaud M, Miralbell R. Patient positioning in prostate radiotherapy: is prone better than supine? Int J Radiat Oncol Biol Phys 2000; 47: van Herk M, Bruce AM, Kroes APG, Shouman T, Touw A, Lebesque JV. Quantification of organ motion during conformal radiotherapy of the prostate by three dimensional image registration. Int J Radiat Oncol Biol Phys 1995; 33: Stroom JC, Koper PCM, Korevaaar GA et al. Internal organ motion in prostate cancer patients treated in prone and supine treatment position. Radiother Oncol 1999; 51: Fokdal L, Honore H, Hoyer M, Meldgaard P, Fode K, von der Maase H. Impact of changes in bladder and rectal filling volume on organ motion and dose distribution of the bladder in radiotherapy for urinary bladder cancer. Int J Radiat Oncol Biol Phys 2004; 59: Muren LP, Ekerold R, Kvinnsland Y, Karlsdottir A, Dahl O. On the use of margins for geometrical uncertainties around the rectum in radiotherapy planning. Radiother Oncol 2004; 70: Wong JR, Grimm L, Uematsu M et al. Image-guided radiotherapy for prostate cancer by CT-linear accelerator combination: prostate movements and dosimetric considerations. Int J Radiat Oncol Biol Phys 2005; 61: Remeijer P, Sonke J-J, Dwarswaard M, van Herk M. Abstract. In The 8th International Workshop on Electronic Portal Imaging. Brighton, London, 2004; Sorcini B, Tilikidis A. Clinical application of image-guided radiotherapy, IGRT (on the Varian OBI platform). Cancer Radiother 2006; 10: Chandra A, Dong L, Huang E et al. Experience of ultrasound-based daily prostate localisation. Int J Radiat Oncol Biol Phys 2003; 56: