Comparison of daily megavoltage electronic portal imaging or kilovoltage imaging with marker seeds to ultrasound imaging or skin marks for prostate localization and treatment positioning in patients with prostate cancer

Int. J. Radiation Oncology Biol. Phys., Vol. 65, No. 5, pp. 1585–1592, 2006 Copyright © 2006 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/06/$–see front matter

doi:10.1016/j.ijrobp.2006.04.019

PHYSICS CONTRIBUTION

COMPARISON OF DAILY MEGAVOLTAGE ELECTRONIC PORTAL IMAGING OR KILOVOLTAGE IMAGING WITH MARKER SEEDS TO ULTRASOUND IMAGING OR SKIN MARKS FOR PROSTATE LOCALIZATION AND TREATMENT POSITIONING IN PATIENTS WITH PROSTATE CANCER CHRISTOPHER F. SERAGO, PH.D.,* STEVEN J. BUSKIRK, M.D.,* TODD C. IGEL, M.D.,† ASHLEY A. GALE, M.S.,* NICOLE E. SERAGO, M.E.,* AND JOHN D. EARLE, M.D.* Departments of *Radiation Oncology and †Urology, Mayo Clinic, Jacksonville, FL Purpose: To compare the accuracy of imaging modalities, immobilization, localization, and positioning techniques in patients with prostate cancer. Methods and Materials: Thirty-five patients with prostate cancer had gold marker seeds implanted transrectally and were treated with fractionated radiotherapy. Twenty of the 35 patients had limited immobilization; the remaining had a vacuum-based immobilization. Patient positioning consisted of alignment with lasers to skin marks, ultrasound or kilovoltage X-ray imaging, optical guidance using infrared reflectors, and megavoltage electronic portal imaging (EPI). The variance of each positioning technique was compared to the patient position determined from the pretreatment EPI. Results: With limited immobilization, the average difference between the skin marks’ laser position and EPI pretreatment position is 9.1 ⴞ 5.3 mm, the average difference between the skin marks’ infrared position and EPI pretreatment position is 11.8 ⴞ 7.2 mm, the average difference between the ultrasound position and EPI pretreatment position is 7.0 ⴞ 4.6 mm, the average difference between kV imaging and EPI pretreatment position is 3.5 ⴞ 3.1 mm, and the average intrafraction movement during treatment is 3.4 ⴞ 2.7 mm. For the patients with the vacuum-style immobilization, the average difference between the skin marks’ laser position and EPI pretreatment position is 10.7 ⴞ 4.6 mm, the average difference between kV imaging and EPI pretreatment position is 1.9 ⴞ 1.5 mm, and the average intrafraction movement during treatment is 2.1 ⴞ 1.5 mm. Conclusions: Compared with use of skin marks, ultrasound imaging for positioning provides an increased degree of agreement to EPI-based positioning, though not as favorable as kV imaging fiducial seeds. Intrafraction movement during treatment decreases with improved immobilization. © 2006 Elsevier Inc. Prostate cancer, Localization, Imaging, Fiducial markers, Radiotherapy.

INTRODUCTION

provide feasible pathways toward achieving highly conformal radiation dose distributions. Conformal treatment plans must include sufficient margins around the gross target volume (GTV) or clinical target volume (CTV) to account for external motion of the patient, internal motion of organs of interest, and setup or geometric treatment delivery uncertainties. The International Commission on Radiation Units Reports 50 and 62 (9, 10) describe a methodology to systematically determine appropriate margins for the planning target volume (PTV). The PTV should be of a sufficient size to provide a margin around the CTV. Nevertheless, any method of reducing the PTV to be closer in size to the CTV is desirable. Interfraction and intrafraction movement of the prostate during radiotherapy has been well documented. A review of

Prostate cancer is the most common cancer in males, and 232,000 new cases are estimated for 2005 (1). Externalbeam radiotherapy is one of the options used to treat localized prostate cancer. Local tumor control probability with external-beam radiotherapy has been shown to correlate positively with increasing radiation dose to the prostate (2–5). Accurate and conformal deposition of the radiation dose is required with careful attention to minimizing the dose to surrounding normal tissues to decrease the risk of adverse effects shown to occur at higher doses (3, 6 – 8). Treatment advances such as three-dimensional conformal radiotherapy and intensity-modulated radiotherapy (IMRT) Reprint requests to: Christopher F. Serago, Ph.D., Department of Radiation Oncology, Mayo Clinic, 4500 San Pablo Road, Jacksonville, FL 32224. Tel: (904) 953-0951; Fax: (904) 953-1004; E-mail: [email protected] Presented at the 47th Annual Meeting of the American Society

for Therapeutic Radiology and Oncology (ASTRO), Denver, CO, October 16 –20, 2005. Received Jan 19, 2006, and in revised form April 20, 2006. Accepted for publication April 21, 2006. 1585

1586

I. J. Radiation Oncology

● Biology ● Physics

prostate motion studies by Langen and Jones (11) reported that the prostate shifts position by an average of 2– 6 mm, with maximum single-occurrence displacements of about 7–20 mm. Most of the studies reviewed by Langen and Jones described interfraction prostate movements. Intrafraction prostate movement, which is generally smaller than interfraction movement, was recently reviewed by Ghilezan et al. (12). Knowledge of intrafraction prostate movement has added importance for IMRT treatments because of the extended treatment time. In addition to prostate movement attributable primarily to daily bladder and rectal variations, Malone et al. (13) have observed respiratory-induced prostate movement. Malone also emphasizes the contribution that patient immobilization, or lack thereof, may have on intrafraction movement. Several strategies have been suggested either to minimize prostate motion or to detect and correct for daily variation of the prostate position. Some have employed the use of rectal probes (14, 15) for the purpose of stabilizing the position of the rectum, thus reducing potential movement of the prostate. Skin-mounted infrared reflectors have been suggested (16) as an improved positioning method compared to conventional skin marks and laser positioning. The use of imaging or imageguided radiotherapy to determine the position of the prostate at the time of treatment is an attractive method gaining greater interest. Several investigators (17–19) have employed an image-guided radiotherapy technique using small implanted radiopaque markers into the prostate. These investigators imaged the markers using megavoltage X-rays with either film or electronic detectors. A computed tomography (CT) scanner installed into the same room as the treatment linear accelerator has been used (20, 21) to obtain daily CT images of the prostate. Similarly, a tomotherapy unit that incorporates directly into the treatment unit a CT imaging device using megavoltage X-rays (22–24) has been implemented. Use of a flat-panel electronic imaging device to acquire a CT dataset from a cone-beam LINAC X-ray source has also been described (25, 26) as a technique to obtain daily three-dimensional image information. Stereoscopic views using kilovoltage X-rays have been employed (27–29) to track target position. Yet another alternative is the use of ultrasound imaging for prostate localization and positioning (30, 31). Our institution has been using daily ultrasound imaging for positioning prostate patients since 2000, and we have previously reported our initial experience (32). Since then, two studies (33, 34) compared ultrasound imaging to electronic portal imaging (EPI) of implanted fiducial markers. These studies reported differences between ultrasoundbased prostate localization and positioning and EPI-based localization and positioning. This study compares daily megavoltage electronic portal imaging of gold marker seeds, kilovoltage X-ray imaging of gold marker seeds, ultrasound imaging, infrared skin-mounted reflectors, and skin marks for localization and positioning patients with prostate cancer. Two different immobilization

Volume 65, Number 5, 2006

techniques were used. Measurement of intrafraction movement of the prostate was also evaluated. METHODS AND MATERIALS Initially, 20 patients with prostate cancer treated at Mayo Clinic in Jacksonville, FL, were retrospectively reviewed (Group I). This retrospective review has received Institutional Review Board approval. The Group I patients underwent treatment with limited immobilization between July 2004 and July 2005. Not all imaging or positioning modalities were used each day to limit the total procedure time. Subsequently, between June 2005 and January 2006, a second group (Group II) of 15 patients underwent treatment with a vacuum device immobilization (BodyFIX; Medical Intelligence, Schwabmuchen, Germany).

Fiducial seed placement In all 35 patients, gold fiducial marker seeds were implanted into the prostate before a treatment planning CT scan. The seed dimensions were 3 mm in length with diameters of 0.8, 1.0, or 1.2 mm. The choice of seed diameter increased over time because of our difficulty visualizing the smaller diameter seeds on the EPI. To visualize the small seeds, the image contrast and brightness often needed to be adjusted, which negatively affected the procedure time. The prostates were implanted with three to six gold seeds; most received either four or five seeds. Although three seeds were sufficient for localization, occasionally seeds were implanted very close to one another, or more rarely a seed migrated, so the presence of extra seeds was beneficial. Typically, the seeds were placed with two at the base of the prostate on the right and left sides; two at the apex of the prostate on the right and left sides; and one seed at the prostate midline. The seeds were inserted transrectally by an urologist using transrectal ultrasound guidance.

Treatment planning Treatment planning CT scans were performed after placement of the fiducial seeds. Patients were supine on a flat carbon-fiber couch top for the scan. For the initial group of 20 patients, limited immobilization consisted of a cylindrical roll under their knees and a wedge between their feet, with their feet secured to the wedge for both the scan and subsequent treatments. The second group of 15 patients were immobilized using the BodyFIX vacuum device shown in Fig. 1. Group I patients had skin-mounted infrared reflectors in place on the abdomen for the CT scans. The locations of these reflectors were permanently marked with tattoos so the position could be reproduced on subsequent treatment days. The CT scan protocol for both groups used a slice thickness of 1.25 mm. Rectal and intravenous contrast was used to enhance the delineation of the bladder and rectum. Patients were scanned and subsequently treated with a full bladder. No bowel preparation was performed before the scan. The range of the CT scan started superiorly at approximately the location of the iliac crests to about 5 cm inferior to the ischial tuberosities. Contours used for planning were drawn by a radiation oncologist on a Pinnacle treatment planning system (Pinnacle; Philips, Milpitas, CA) using the CT scans from the treatment planning session. Contours of interest included the prostate GTV, seminal vesicles, bladder, and rectum. For Group II, the PTV was determined by adding a uniform margin of 5 mm around the GTV in all directions with the exception that the anterior rectal wall was used as a limiting boundary. The PTV for Group I patients was slightly

Comparison of imaging for prostate patients

● C. F. SERAGO et al.

1587

Daily positioning steps

Fig. 1. BodyFIX immobilization device.

larger because of expansion of the GTV by 8 mm in the lateral direction. No separate CTV was used in the planning process. The location of the treatment isocenter was the approximate geometric center of the prostate GTV. Right, left, and anterior tattoo marks were made on the patient’s skin to correspond to the treatment isocenter. The prostate contours and the prostate’s geometric isocenter position were exported from the planning system to both the B-mode Acquisition and Targeting (BAT) ultrasound system (Nomos, Sewickley, PA) and the Exactrac kilovoltage imaging system (BrainLAB Inc., Heimstetten, Germany). Also exported from the planning system to the Exactrac system were the axial CT images on which the gold seed positions were digitized. Digitally reconstructed radiographs (DRRs) with the gold seeds contoured were exported from the planning system to the PortalVision EPI (Varian Medical Systems, Palo Alto, CA) electronic portal imaging system. Both the BrainLAB Exactrac and Varian PortalVision software programs register DRRs with gold seeds and isocenter geometric positions derived from the planning CT scan to daily acquired images to calculate translational and rotational corrections. For this study, only translational corrections were done. Each patient had an IMRT plan, usually with seven coplanar 6-MV beams. The prescribed dose for the Group I patients ranged from 72.0 to 77.4 Gy (mean, 75.3 Gy) in 40 – 43 fractions, whereas the dose for the Group II patients ranged from 77.4 to 79.2 Gy (mean, 79.1 Gy) in 43 or 44 fractions. Anteroposterior (AP) and right lateral beams shaped to conform to the prostate GTV with a margin were an integral part of the plan to be used for pretreatment and posttreatment EPIs. The approximate treatment time for the seven IMRT fields was 5– 6 min. The dose from the daily EPIs was incorporated in the treatment plan with the daily dose from these images recorded as part of the total dose. The locations of the fiducial seeds were contoured to facilitate observation of their position on subsequent DRRs. The AP and lateral DRRs with the highlighted fiducial seeds formed the basis for prostate position confirmation during each treatment day. If the corresponding pretreatment AP and lateral EPI fiducial seed locations relative to the beam aperture agreed with the DRR, then it was assumed that the prostate gland was in the correct position at that time. All other measurements of position at different times are referenced to the position as determined from the daily pretreatment set of EPIs.

The daily position of the prostate for the Group I patients was set and adjusted with five methods: (1) orthogonal laser lines projecting toward the isocenter used with skin tattoo marks; (2) skin-mounted infrared reflectors; (3) ultrasound imaging; (4) kilovoltage stereoscopic electronic X-ray images; and (5) megavoltage EPI. Neither skin-mounted infrared reflectors nor ultrasound imaging was used for the Group II patients. We made this choice because the results of the Group I patients indicated these were less accurate positioning methods, and the immobilization device used for the Group II patients would hinder those methods. Another difference between the groups was the method of immobilization. The Group I patients had limited immobilization, whereas the Group II patients had BodyFIX immobilization. The daily sequence of positioning steps is given in the following paragraphs. Step 1 on each treatment day for both Group I and II was to align the patient to three skin marks using an orthogonal set of lasers. The skin marks, actually small tattoos on the patient’s AP and right and left lateral sides, indicated the approximate center of the prostate GTV, and are verified before the initial treatment with orthogonal X-rays using a conventional radiographic simulator. The patient treatment couch position coordinates were recorded after the skin mark to laser alignment was completed each day. Step 2 alternated each day for Group I patients between: (1) BAT ultrasound positioning and (2) Exactrac infrared skin reflector positioning followed immediately by Exactrac kV imaging. BAT ultrasound positioning on Tuesdays and Thursdays consisted of obtaining axial and sagittal ultrasound images. The isocenter position and the contours of the prostate, bladder, and rectum exported from the treatment planning system were used for patient positioning. This process has been described previously (30, 31). The patient treatment couch position coordinates were recorded after the BAT patient positioning was completed each day. Group II patients did not have BAT positioning performed. Steps 2A and 2B on Monday, Wednesday, and Friday for Group I patients consisted of Exactrac positioning with use of the skin mounted infrared reflectors (Step 2A), followed by use of the kilovoltage stereoscopic X-ray imaging and patient positioning (Step 2B). A description of the Exactrac imaging and positioning system has been given previously (35). The patient treatment couch positions were recorded after Steps 2A and 2B. The version of Exactrac in use for the Group I patients had a single image detector mounted under the patient couch with two ceiling-mounted X-ray tubes. The Exactrac version in use for the Group II patients had dual ceiling-mounted image detectors with the X-ray tubes installed in the floor (see Fig.2). Group II patients had only Exactrac X-ray positioning for Step 2; the skin-mounted infrared reflectors were not used. Step 3 just before treatment each day for both Groups I and II was to obtain an orthogonal pair of AP and lateral EPIs displaying the gold fiducial seeds in the prostate. The PortalVision software has tools to compare the location of the gold seeds with respect to the treatment planning CT simulation reference digitally reconstructed radiograph. This procedure has been described by Herman et al. (17). The patient couch position was recorded subsequent to EPI positioning. Step 4 each day for all patients was to treat the patient. Step 5 each day for both Groups I and II was a repetition of the third step (EPI localization and positioning) immediately after completion of treatment. For a final time, the coordinates of the couch position were determined and recorded.

1588

I. J. Radiation Oncology

● Biology ● Physics

Volume 65, Number 5, 2006

Table 1. Task time for each positioning procedure steps

RESULTS

Step

Positioning procedure

Group

Time (min)

1 2 2A 2B 2 3 4 5

Align lasers to skin tattoos BAT ultrasound Exactrac infrared skin reflectors Exactrac kV X-ray, single detector Exactrac kV X-ray, dual detector PortalVision MV EPI Treatment 7 IMRT beams PortalVision MV EPI

I and II I I I II I and II I and II I and II

2 3–4 1 3–4 2 3–4 5–6 1–2

The patient couch position, as determined by the pretreatment set of EPI images, was assumed to place the prostate in the correct position for treatment, and all other couch positions determined at their imaging and positioning times were compared with the pretreatment EPI position. Displacement differences and absolute values of the couch position differences were calculated. Table 2 summarizes the average differences of couch position for the 20 Group I patients for each positioning modality. The total number of measurements for each comparison is listed in the rightmost column. Differences in the total number of measurements is attributable to a variety of factors including: BAT and Exactrac positioning on alternating days, equipment malfunctions, poor ultrasound image quality, or patient discomfort causing an abbreviated session. The average vertical, lateral, longitudinal, total absolute differences, and displacement differences are given in Table 2, as well as the standard deviations of the total

Abbreviations: EPI ⫽ electronic portal imaging; kV ⫽ kilovoltage; IMRT ⫽ intensity-modulated radiotherapy; BAT ⫽ B-mode acquisition and targeting; MV ⫽ megavoltage.

Each of the localization and positioning steps takes time. During this time, of course, either motion of the patient’s body or internal organ movement within the patient was possible. The average times required to perform each of the positioning steps are summarized in Table 1.

Table 2. Average couch position absolute differences and displacements from pretreatment EPI position in all Group I patients with limited immobilization, and differences between positioning modality averages with intrafraction movement Group I: averages for all patients Limited immobilization Absolute value differences (mm)

Item A EPI vs. skin/lasers Item B EPI vs. Exactrac skin infrared Item C EPI vs. BAT ultrasound Item D EPI vs. Exactrac kV imaging Item E EPI before vs. EPI after Rx (intrafraction movement)

VERT

LAT

LONG

TOTAL*

No.

5.4 5.4 3.6 1.6 1.7

3.7 6.4 4.0 1.8 1.7

3.9 5.9 2.0 1.0 1.2

9.1 ⫾ 5.3 11.8 ⫾ 7.2 7.0 ⫾ 4.6 3.5 ⫾ 3.1 3.4 ⫾ 2.7

815 231 322 237 771

Group I: averages for all patients limited immobilization Displacement (mm)

EPI EPI EPI EPI EPI

vs. skin/lasers vs. Exactrac skin infrared vs. BAT ultrasound vs. Exactrac kV imaging before vs. EPI after Rx

VERT*

LAT*

LONG*

No.

–3.0 ⫾ 6.0 –3.8 ⫾ 5.1 –1.4 ⫾ 4.8 –0.7 ⫾ 2.4 0.6 ⫾ 2.4

0.1 ⫾ 6.4 –0.2 ⫾ 9.4 –0.4 ⫾ 3.3 –0.4 ⫾ 2.3 –0.1 ⫾ 2.3

3.3 ⫾ 5.0 3.9 ⫾ 7.2 1.5 ⫾ 5.7 –0.7 ⫾ 3.0 –0.6 ⫾ 2.7

815 231 322 237 771

Group I: positioning modality differences from intrafraction movement

Skin/lasers (items A–E) Exactrac skin infrared (items B–E) BAT ultrasound (items C–E) Exactrac kV imaging (Items D–E)

Difference

95% CI

5.7 mm 8.4 mm 3.6 mm 0.1 mm

5.3⫺6.1 7.5⫺9.3 3.1⫺4.1 ⫺0.3⫺0.5

Abbreviations: EPI ⫽ electronic portal imaging; kV ⫽ kilovoltage; IMRT ⫽ intensity-modulated radiotherapy; CI ⫽ confidence interval; BAT ⫽ B-mode acquisition and targeting; LAT ⫽ lateral; LONG ⫽ longitudinal; VERT ⫽ vertical; Rx ⫽ radiotherapy. * Values are mean ⫾ standard deviation.

Comparison of imaging for prostate patients

● C. F. SERAGO et al.

1589

Table 3. Average couch position absolute differences and displacements from pretreatment EPI position for all Group II patients with vacuum-style immobilization, and differences between positioning modality averages with intrafraction movement Group II: averages for all patients BodyFIX immobilization Absolute value differences (mm)

Item A EPI vs. skin/lasers Item B EPI vs. Exactrac kV imaging Item C EPI before vs. EPI after Rx (intrafraction movement)

VERT

LAT

LONG

TOTAL*

No.

7.3 0.9 1.2

4.1 0.9 0.9

3.5 0.7 0.7

10.7 ⫾ 4.6 1.9 ⫾ 1.5 2.1 ⫾ 1.5

551 520 557

Group II: averages for all patients BodyFIX immobilization Displacement (mm)

EPI vs. skin/lasers EPI vs. exactrac kV imaging EPI before vs. EPI after Rx

VERT*

LAT*

LONG*

No.

1.8 ⫾ 8.0 0.2 ⫾ 1.4 0.5 ⫾ 1.8

–0.1 ⫾ 5.2 0.1 ⫾ 1.2 –0.2 ⫾ 1.0

1.5 ⫾ 6.3 –0.1 ⫾ 1.4 0.0 ⫾ 1.5

551 520 557

Group II: positioning modality differences from intrafraction movement

Skin/lasers (Items A–C) Exactrac kV imaging (Items B–C)

Difference

95% CI

8.6 mm –0.2 mm

8.2–9.0 –0.4–0.0

Abbreviations: EPI ⫽ electronic portal imaging; kV ⫽ kilovoltage; IMRT ⫽ intensity-modulated radiotherapy; CI ⫽ confidence interval; Rx - radiotherapy; LAT ⫽ lateral; LONG ⫽ longitudinal; VERT ⫽ vertical. * Values are mean ⫾ standard deviation.

difference and standard deviations of the average displacements. In Table 2, the EPI before vs. EPI after treatment average total difference and standard deviation, 3.4 ⫾ 2.7 mm, was an indication of the average intrafraction movement of the prostate for this group of patients. This movement could be attributed to either internal organ or external patient motion. Because each positioning and imaging step occurred at a separate point in time, it is possible that the differences in prostate position were not entirely the result of different positioning modalities, but could be caused at least in part by intrafraction movement. Thus in the lower section of Table 2, total intrafraction movement (E: EPI before vs. EPI after) was subtracted from the total differences (Item A–Item E, Table 2) of couch (prostate) position for each positioning and imaging modality. The 95% confidence interval for the difference between the averages was also calculated to demonstrate whether the differences between positioning modalities were significant after the observed intrafraction movement was removed. For Group I, the positioning modality difference between positioning with Exactrac kV imaging and the total average intrafraction movement was 0.1 mm with a 95% confidence interval of ⫺0.3 to 0.5 mm (not statistically different). This is shown in the bottom line of Table 2 (Item D–Item E). The positioning modality difference between positioning with

BAT ultrasound and the total average intrafraction movement was 3.6 mm with of 95% confidence interval of 3.1– 4.1 mm, which was less than either the skin/laser or Exactrac skin infrared marker modality difference from intrafraction movement. The positioning modality difference between skin/laser positioning and the total average intrafraction movement was 5.7 mm with a 95% confidence interval of 5.3– 6.1 mm, and the positioning modality difference between Exactrac skin infrared markers and the total average intrafraction movement was 8.4 mm with a 95% confidence interval of 7.5–9.3 mm. Table 3, similar to Table 2, summarizes the average differences of couch position for the 15 Group II patients. These patients had the BodyFIX vacuum immobilization as opposed to the limited immobilization of Group I, and the BAT ultrasound and Exactrac infrared skin reflector positioning were not done. The skin tattoo marks showed a 10.7 ⫾ 4.6 mm difference in couch position compared with the pretreatment EPI-determined position. The average difference between positioning with kV imaging and the pretreatment EPI-determined position was 1.9 ⫾ 1.5 mm. Also shown in the central portion of Table 3 are average displacement differences with standard deviations. For Group II, the positioning modality difference between Exactrac kV imaging and the total average intrafraction movement was ⫺0.2 mm with a 95% confidence in-

1590

I. J. Radiation Oncology

● Biology ● Physics

terval of ⫺0.4 to 0.0 mm; as with Group I, not statistically different. The positioning modality difference between skin/ laser positioning and the total average intrafraction movement was 8.6 mm with 95% confidence interval of 8.2–9.0 mm. Of the 771 daily intrafraction move measurements from the Group I, 3.1% (24/771) had intrafraction moves greater than 1 cm, whereas for Group II with improved immobilization, 0.5% (3/557) had greater than 1 cm intrafraction moves. The difference of the total average intrafraction move of Group I compared with Group II is 1.3 mm with a confidence interval of 1.1–1.5 mm. Seed migration was noted in 2 of the total 35 combined Group I and II patients. In each of these patients, a single seed moved more than 3 mm. Each of these seeds had been implanted at the very periphery of the prostate or possibly in periprostatic tissue. DISCUSSION The positioning, localization, and imaging modalities described in this article each have both favorable and unfavorable competing arguments for utilization. Several factors such as equipment cost, procedure time, procedure complexity, patient comfort, invasiveness of procedure, staff training, and staff expertise should be considered. Skin tattoo marks used with orthogonal lasers, for example, are the simplest and most common historical method, yet are also well documented to not be reliable, especially for an organ such as the prostate that is known to exhibit internal motion. In fact, prior demonstration that skin marks are not reliable indicators of prostate position was an incentive for development of image-guided techniques. As mentioned previously, the review article by Langen and Jones (11) summarizes many prostate organ motion studies showing mean prostate interfraction displacements of 2– 6 mm with maximum single-day displacements as large as 20 mm. Use of other skin-based markers, such as the skin-mounted infrared reflectors, can be expected to have similar shortcomings as the skin tattoos. Weiss et al. (16) have reported that the Exactrac skin infrared reflectors provide similar accuracy to the skin tattoo/laser positioning, and our results support this finding as well. Again, because of the internal movement of the prostate, that is not surprising. It is interesting to note that the skin/tattoo positioning accuracy with the vacuum immobilization for Group II patients is not improved compared with the limited immobilization Group I patients. We believe this is because the vacuum device interferes somewhat with visualization of the skin tattoos, and that the therapists over time may have relaxed their tolerance of setup to the skin marks having gained confidence in the daily imaging. Ultrasound image guidance is an attractive option because it is noninvasive compared with implantation of gold marker seeds. However, two publications in 2003 (33, 34) that compare ultrasound imaging with megavoltage imaging of marker seeds reported limited positioning improvement

Volume 65, Number 5, 2006

using daily ultrasound imaging. These reports suggest that interuser variability, CT and ultrasound contour differences, and intrafraction organ motion may be some of the factors contributing to an unfavorable correlation of ultrasound and megavoltage images. Other potential problems with ultrasound imaging are the quality of the image may be inadequate and that radiation therapists performing the ultrasound image acquisition may not be experienced ultrasonographers. Yet another consideration is the possible appearance differences between a CT and an ultrasound image of the prostate and surrounding anatomy (36). In a previous study from our institution (32), we found that in patients selected for ultrasound positioning, 18 –32% did not have adequate image quality. Although this rejection rate is higher than other reports (37), some fraction of patients will not be good ultrasound imaging candidates. Nonetheless, this study demonstrates that BAT ultrasound imaging is a better method of positioning than using skin markers alone, although it is not as good as daily imaging of implanted marker seeds. Megavoltage EPI and Exactrac kilovoltage X-ray positioning appear to provide equivalent results. Our results showed the average difference between megavoltage EPI and Exactrac X-ray positioning is 3.5 mm for patients with limited immobilization and 1.9 mm average difference for BodyFIX immobilized patients. These values are nearly the same magnitude as the measured intrafraction movement.

Fig. 2. (a) Exactrac single detector, 2 X-ray tube configuration. (b) Exactrac dual detector, 2 X-ray tube configuration.

Comparison of imaging for prostate patients

There are two advantages of the present version of the Exactrac imaging/positioning system compared with use of PortalVision EPI. First, the daily dose from the EPIs of 5–7 cGy per pair of orthogonal EPIs is much larger than using kilovoltage X-rays. Second, use of kilovoltage X-rays for imaging provides much better visualization of the gold marker seeds compared with megavoltage X-rays, especially the lateral EPI view. In addition, the Exactrac system is capable of auto-couch movement, based on its calculated positioning correction. The present version of PortalVision EPI requires the therapist to manually move the couch. An important assumption implicit in the validity of positioning based on marker seed locations within the prostate is that the seeds neither migrate nor does the prostate itself deform significantly. Schallenkamp et al. (38) discussed this thoroughly and data from their study of seed migration in 20 prostate patients showed average intermarker movement of less than 1 mm. Also of interest is the decreased intrafraction movement observed when using the BodyFIX immobilization device. We report in this study average intrafraction movements of 3.4 ⫾ 2.7 mm for patients with limited immobilization and 2.1 ⫾ 1.5 mm for patients with the BodyFIX device. Although this study did not attempt to separate internal organ motion from external body motion, we believe it is likely that this decrease of intrafraction movement reflects less external body motion. Our results of intrafraction movement may be compared to a recent review of the literature of intrafraction prostate motion studies published by Ghilezan et al. (12). Most of the studies Ghilezan reviewed reported their intrafraction movements as displacements (⫾ standard deviation) that

● C. F. SERAGO et al.

1591

have an average near zero, rather than absolute distances. Ghilezan further notes the wide variety of methodology of the motion studies including such differences such as: imaging (ultrasound, CT, MRI, fluoroscopy), measurement time (10 s to 20 min), patient position (supine, prone), immobilization technique, and presence or absence of fiducial markers, all of which can make a direct comparison difficult. Nonetheless, one can compare standard deviations as a measure of intrafraction movement and absolute differences when available. With those caveats, our data show similar results for intrafraction movement to Mah et al. (39) and Padhani et al. (40), but are slightly larger than those reported by others (41– 44). The combination of IMRT and daily image guidance has had a clinical impact on our practice, primarily seen by the evidence of dose escalation. As noted previously, the mean dose of the Group I patients was 75.3 Gy compared with 79.1 Gy for the subsequently treated Group II patients. Additionally, the results from the analysis of the Group I patients allowed us to decrease our PTVs for the Group II patients. CONCLUSIONS If one assumes that pretreatment EPI-determined positioning has placed the patient in the correct location, then we can conclude that positioning techniques that rely on skin marks for patients with prostate cancer are inferior to all the image-guided techniques investigated in this study. Further, we conclude that both megavoltage and kilovoltage imaging have similar accuracy, and are better than ultrasound imaging. Patient immobilization may be improved using a device such as the one described in this report.

REFERENCES 1. Cancer facts and figures 2005. Dallas, TX: American Cancer Society, 2005. 2. Zietman AL, DeSilvio ML, Slater JD, et al. Comparison of conventional-dose vs high-dose conformal radiation therapy in clinically localized adenocarcinoma of the prostate: a randomized controlled trial [see comment]. JAMA 2005;294:1233–1239. 3. Hanks GE, Hanlon AL, Schultheiss TE, et al. Dose escalation with 3D conformal treatment: Five year outcomes, treatment optimization, and future directions. Int J Radiat Oncol Biol Phys 1998;41:501–510. 4. Zelefsky MJ, Fuks Z, Hunt M, et al. High dose radiation delivered by intensity modulated conformal radiotherapy improves the outcome of localized prostate cancer [see comment] [erratum appears in J Urol 2001 Nov;166(5):1839]. J Urol 2001;166:876 – 881. 5. Pollack A, Zagars GK, Smith LG, et al. Preliminary results of a randomized radiotherapy dose-escalation study comparing 70 Gy with 78 Gy for prostate cancer. J Clin Oncol 2000;18: 3904 –3911. 6. Michalski JM, Purdy JA, Winter K, et al. Preliminary report of toxicity following 3D radiation therapy for prostate cancer on 3DOG/RTOG 9406. Int J Radiat Oncol Biol Phys 2000;46: 391– 402. 7. Skwarchuk MW, Jackson A, Zelefsky MJ, et al. Late rectal toxicity after conformal radiotherapy of prostate cancer (I):

8.

9. 10.

11. 12. 13. 14.

15.

Multivariate analysis and dose-response. Int J Radiat Oncol Biol Phys 2000;47:103–113. Hartford AC, Niemierko A, Adams JA, et al. Conformal irradiation of the prostate: estimating long-term rectal bleeding risk using dose-volume histograms. Int J Radiat Oncol Biol Phys 1996;36:721–730. ICRU. ICRU report 50: Prescribing, recording and reporting photon beam therapy. Bethesda: International Commission on Radiation Units and Measurements; 1993. ICRU. ICRU report 62: Prescribing, recording and reporting photon beam therapy. Supplement to ICRU Report 5. Bethesda: International Commission on Radiation Units and Measurements;1999. Langen KM, Jones DT. Organ motion and its management. Int J Radiat Oncol Biol Phys 2001;50:265–278. Ghilezan MJ, Jaffray DA, Siewerdsen JH, et al. Prostate gland motion assessed with cine-magnetic resonance imaging (cineMRI). Int J Radiat Oncol Biol Phys 2005;62:406 – 417. Malone S, Crook JM, Kendal WS, et al. Respiratory-induced prostate motion: quantification and characterization. Int J Radiat Oncol Biol Phys 2000;48:105–109. Benk VA, Adams JA, Shipley WU, et al. Late rectal bleeding following combined X-ray and proton high dose irradiation for patients with stages T3-T4 prostate carcinoma. Int J Radiat Oncol Biol Phys 1993;26:551–557. Teh BS, Mai WY, Uhl BM, et al. Intensity-modulated radia-

1592

16.

17.

18.

19.

20.

21.

22. 23. 24. 25. 26. 27. 28.

29.

30.

I. J. Radiation Oncology

● Biology ● Physics

tion therapy (IMRT) for prostate cancer with the use of a rectal balloon for prostate immobilization: Acute toxicity and dose-volume analysis. Int J Radiat Oncol Biol Phys 2001;49:705–712. Weiss E, Vorwerk H, Richter S, et al. Interfractional and intrafractional accuracy during radiotherapy of gynecologic carcinomas: A comprehensive evaluation using the ExacTrac system. Int J Radiat Oncol Biol Phys 2003;56:69 –79. Herman MG, Pisansky TM, Kruse JJ, et al. Technical aspects of daily online positioning of the prostate for three-dimensional conformal radiotherapy using an electronic portal imaging device. Int J Radiat Oncol Biol Phys 2003;57: 1131–1140. Balter JM, Sandler HM, Lam K, et al. Measurement of prostate movement over the course of routine radiotherapy using implanted markers. Int J Radiat Oncol Biol Phys 1995;31: 113–118. Alasti H, Petric MP, Catton CN, et al. Portal imaging for evaluation of daily on-line setup errors and off-line organ motion during conformal irradiation of carcinoma of the prostate. Int J Radiat Oncol Biol Phys 2001;49:869 – 884. Shiu AS, Chang EL, Ye JS, et al. Near simultaneous computed tomography image-guided stereotactic spinal radiotherapy: An emerging paradigm for achieving true stereotaxy. Int J Radiat Oncol Biol Phys 2003;57:605– 613. 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;61:561–569. Mackie TR, Balog J, Ruchala K, et al. Tomotherapy. Semin Radiat Oncol 1999;9:108 –117. Mackie TR, Kapatoes J, Ruchala K, et al. Image guidance for precise conformal radiotherapy. Int J Radiat Oncol Biol Phys 2003;56:89 –105. Langen KM, Zhang Y, Andrews RD, et al. Initial experience with megavoltage (MV) CT guidance for daily prostate alignments. Int J Radiat Oncol Biol Phys 2005;62:1517–1524. Pouliot J, Bani-Hashemi A, Chen J, et al. Low-dose megavoltage cone-beam CT for radiation therapy. Int J Radiat Oncol Biol Phys 2005;61:552–560. Jaffray DA, Siewerdsen JH, Wong JW, et al. Flat-panel conebeam computed tomography for image-guided radiation therapy. Int J Radiat Oncol Biol Phys 2002;53:1337–1349. Murphy MJ. Tracking moving organs in real time. Semin Radiat Oncol 2004;14:91–100. Rietzel E, Rosenthal SJ, Gierga DP, et al. Moving targets: Detection and tracking of internal organ motion for treatment planning and patient set-up. Radiother Oncol 2004;73(Suppl 2):S68 –S72. Shirato H, Shimizu S, Kitamura K, et al. Four-dimensional treatment planning and fluoroscopic real-time tumor tracking radiotherapy for moving tumor. Int J Radiat Oncol Biol Phys 2000;48:435– 442. Lattanzi J, McNeeley S, Pinover W, et al. A comparison of daily CT localization to a daily ultrasound-based system in

Volume 65, Number 5, 2006

31.

32.

33.

34. 35. 36.

37.

38.

39. 40.

41. 42.

43.

44.

prostate cancer [see comment]. Int J Radiat Oncol Biol Phys 1999;43:719 –725. Mohan D, Kupelian P, Willoughby M. Short-course intensitymodulated radiotherapy for localized prostate cancer with daily transabdominal ultrasound localization of the prostate gland. Int J Radiat Oncol Biol Phys 2000;46:575–580. Serago CF, Chungbin SJ, Buskirk SJ, et al. Initial experience with ultrasound localization for positioning prostate cancer patients for external beam radiotherapy. Int J Radiat Oncol Biol Phys 2002;53:1130 –1138. Van den Heuvel F, Powell T, Seppi E, et al. Independent verification of ultrasound based image-guided radiation treatment, using electronic portal imaging and implanted gold markers. Med Phys 2003;30:2878 –2887. Langen KM, Pouliot J, Anezinos C, et al. Evaluation of ultrasound-based prostate localization for image-guided radiotherapy. Int J Radiat Oncol Biol Phys 2003;57:635– 644. Yan H, Yin FF, Kim JH. A phantom study on the positioning accuracy of the Novalis Body system. Med Phys 2003;30: 3052–3060. Narayana V, Roberson PL, Winfield RJ, et al. Impact of ultrasound and computed tomography prostate volume registration on evaluation of permanent prostate implants. Int J Radiat Oncol Biol Phys 1997;39:341–346. Willoughby T, Weinhous MS, Kupelian PA. Daily transabdominal ultrasound localization of the prostate gland with the BAT system. Int J Radiat Oncol Biol Phys 2000;48(Suppl.): 308 –309. Schallenkamp JM, Herman MG, Kruse JJ, et al. Prostate position relative to pelvic bony anatomy based on intraprostatic gold markers and electronic portal imaging. Int J Radiat Oncol Biol Phys 2005;63:800 – 811. Mah D, Steckner M, Palacio E, et al. Measurement of intrafractional prostate motion using magnetic resonance imaging. Int J Radiat Oncol Biol Phys 2002;54:568 –575. Padhani AR, Khoo VS, Suckling J, et al. Evaluating the effect of rectal distension and rectal movement on prostate gland position using cine MRI. Int J Radiat Oncol Biol Phys 1999; 44:525–533. Huang E, Dong L, Chandra A, et al. Intrafraction prostate motion during IMRT for prostate cancer. Int J Radiat Oncol Biol Phys 2002;53:261–268. Kitamura K, Shirato H, Seppenwoolde Y, et al. Three-dimensional intrafractional movement of prostate measured during real-time tumor-tracking radiotherapy in supine and prone treatment positions. Int J Radiat Oncol Biol Phys 2002;53: 1117–1123. Nederveen AJ, van der Heide UA, Dehnad H, et al. Measurements and clinical consequences of prostate motion during a radiotherapy fraction. Int J Radiat Oncol Biol Phys 2002;53: 206 –214. Shimizu S, Shirato H, Kitamura K, et al. Use of an implanted marker and real-time tracking of the marker for the positioning of prostate and bladder cancers. Int J Radiat Oncol Biol Phys 2000;48:1591–1597.