VAGINAL MOTION AND BLADDER AND RECTAL VOLUMES DURING PELVIC INTENSITY-MODULATED RADIATION THERAPY AFTER HYSTERECTOMY

Transcription

1 doi: /j.ijrobp Int. J. Radiation Oncology Biol. Phys., Vol. 82, No. 1, pp , 2012 Copyright Ó 2012 Elsevier Inc. Printed in the USA. All rights reserved /$ - see front matter CLINICAL INVESTIGATION Gynecologic Cancer VAGINAL MOTION AND BLADDER AND RECTAL VOLUMES DURING PELVIC INTENSITY-MODULATED RADIATION THERAPY AFTER HYSTERECTOMY ANUJA JHINGRAN, M.D.,* MOHAMMAD SALEHPOUR, PH.D., y MARIANNE SAM, B.S.,* LARRY LEVY, M.S.,* AND PATRICIA J. EIFEL, M.D.* Departments of *Radiation Oncology and y Radiation Physics, The University of Texas M. D. Anderson Cancer Center, Houston, TX Purpose: To evaluate variations in bladder and rectal volume and the position of the vaginal vault during a 5-week course of pelvic intensity-modulated radiation therapy (IMRT) after hysterectomy. Methods and Materials: Twenty-four patients were instructed how to fill their bladders before simulation and treatment. These patients underwent computed tomography simulations with full and empty bladders and then underwent rescanning twice weekly during IMRT; patients were asked to have full bladder for treatment. Bladder and rectal volumes and the positions of vaginal fiducial markers were determined, and changes in volume and position were calculated. Results: The mean full and empty bladder volumes at simulation were 480 cc (range, 122 1,052) and 155 cc (range, ), respectively. Bladder volumes varied widely during IMRT: the median difference between the maximum and minimum volumes was 247 cc (range, ). Variations in rectal volume during IMRT were less pronounced. For the 16 patients with vaginal fiducial markers in place throughout IMRT, the median maximum movement of the markers during IMRT was 0.59 cm in the right left direction (range, 0 0.9), 1.46 cm in the anterior posterior direction (range, ), and 1.2 cm in the superior inferior direction (range, ). Large variations in rectal or bladder volume frequently correlated with significant displacement of the vaginal apex. Conclusion: Although treatment with a full bladder is usually preferred because of greater sparing of small bowel, our data demonstrate that even with detailed instruction, patients are unable to maintain consistent bladder filling. Variations in organ position during IMRT can result in marked changes in the position of the target volume and the volume of small bowel exposed to high doses of radiation. Ó 2012 Elsevier Inc. Intensity-modulated radiation therapy, Gynecologic malignancies, Bladder filling, Organ motion, Vaginal movement. INTRODUCTION Conventional techniques for pelvic radiation therapy after hysterectomy for carcinoma of the uterine cervix or corpus involve either two or four static photon fields. These techniques expose most of the contents of the true pelvis to the prescribed dose (usually Gy in fractions). After hysterectomy, the small bowel tends to fall into the vacated space in the true pelvis, increasing the amount of bowel treated to high doses. Recently, highly conformal techniques such as intensity-modulated radiation therapy (IMRT) have been developed that can reduce the radiation dose delivered to the small bowel during treatment. Preliminary studies suggest that acute and chronic side effects may be reduced when patients with gynecologic malignancies are treated with IMRT rather than with conventional four-field arrangements (1 3). However, the authors of these studies admit that an important limitation of their studies is the absence of data on dayto-day variations in the positions of the abdominal and pelvic organs during irradiation. Without such data, clinicians must rely on clinical judgement to estimate the margin necessary to account for internal organ motion. In an earlier report on patients undergoing IMRT after hysterectomy, we showed that the integral dose delivered to small bowel increased dramatically as the margin to account for possible setup error and organ motion was increased. In that study, we emphasized the importance of accurate target volume definition and effective patient immobilization and the need to develop methods to measure and minimize the uncertainties associated with internal organ motion (1). The vaginal vault and central pelvic tissues included in the clinical target volume (CTV) are immediately adjacent to the bladder and rectum. Before the study that we report here, we had experiences suggesting that the position of the vagina could shift by as much as 2 cm simply with variations in bladder filling. Changes in rectal contents also Reprint requests to: Anuja Jhingran, M.D., Department of Radiation Oncology, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd., Box 212, Houston, TX. Tel: (713) ; Fax: (713) ; ajhingra@mdanderson.org 256 Conflict of interest: none. Received Nov 9, 2009, and in revised form Aug 4, Accepted for publication Aug 17, 2010.

2 Vaginal movement during pelvic IMRT d A. JHINGRAN et al. 257 caused apparent shifts in the target position. To further define the magnitude of changes in target position as a result of bladder and rectal volume, we designed a prospective study in which computed tomography (CT) images were obtained twice weekly during posthysterectomy pelvic IMRT; these images were used to determine variations in the positions of internal organs in patients who were carefully instructed to fill their bladders before simulation and intratreatment CT imaging. MATERIALS AND METHODS Twenty-four women who were referred to the Department of Radiation Oncology for chemoradiation therapy after hysterectomy for carcinoma of the uterine cervix or endometrium with high-risk local findings were entered in a prospective, institutional review board approved study of pelvic IMRT. All patients gave informed consent for participation in the study. Patients were eligible for the study if they had undergone a hysterectomy for carcinoma of the uterine cervix or endometrium and required postoperative radiation or chemoradiation therapy because of positive pelvic lymph nodes or other high-risk factors. Patients were ineligible if they had disease outside the pelvis, mental status changes or bladder control problems making it difficult for the patient to comply with bladder-filling instructions, or body weight or lateral body diameter exceeding the limits of the treatment table or CT scanner. CT simulation Patients were instructed to come to their simulation with a full bladder. Before simulation, all the patients were given written instructions to empty the bladder 60 minutes before simulation and then to drink 24 ounces of liquid during the subsequent 10 minutes, finishing 50 minutes before simulation. Patients were also counseled to follow the same practice before each daily radiation treatment. Sixteen of the 24 patients also had two radiopaque platinum marker seeds inserted in the apex of the vagina before simulation; however, in some one of the seeds fell out before simulation or during treatment. For simulation and treatment, patients were immobilized in a supine position using Vac-Loc devices to constrain the upper and lower body. After acquisition of the initial planning CT scan (Picker 5000; Marconi Medical Systems, Cleveland, OH), the patient was asked to empty her bladder, and a second scan was obtained with the patient s bladder empty. Patients were scanned from the upper abdomen to below the perineum using 2.5-mm CT slices. Critical structures, including the bladder, rectum, and small bowel, were contoured on full and empty bladder scans using the Pinnacle software suite (Philips Radiation Oncology Systems, Milpitas, CA). The entire contents of the bladder and rectum were contoured. The rectum was contoured from the sigmoid flexure to the internal anal sphincter; care was taken to maintain consistency in the proximal and distal limits of rectal contouring on the multiple scans taken for each patient. Vaginal marker seeds were also delineated on both scans. For treatment planning, scans obtained with full and empty bladder were fused, and an internal target volume (ITV) was defined that included the positions of the vagina and paravaginal tissues on both scans. In most cases, the vagina could be readily differentiated from the bladder, urethra, and rectum by viewing the tissue planes on axial, coronal, and sagittal views of the pelvis. The vaginal CTV was contoured to include the vaginal soft tissue between the bladder and rectum and laterally to cover the paravaginal tissues up to the pelvic musculature. However, after the first several cases were entered in the trial, it became apparent that the apical extent of the vagina could not be consistently identified on the serial CT scans; in subsequent cases, platinum marker seeds were inserted in the vaginal apex using a specialized needle equipped with a hub that limited the depth of insertion. The vaginal CTV was usually defined to include these seeds with at least a 1-cm margin superiorly; however, for some patients treated after hysterectomy for cervical cancer, a greater margin was required to cover areas of risk along the margin of resection on the posterior bladder wall. In addition, a CTV was defined that included the regional nodes. Treatment All patients were treated with IMRT to a total dose of 45 Gy to the vaginal ITV plus a 1.0-cm margin and the nodal CTV plus a 7-mm margin. After completion of IMRT, patients received two or three high-dose-rate vaginal cuff treatments of 5 Gy each prescribed to the surface of the upper third of the vagina. Patients were treated in a room equipped with a CT on rails; this equipment allowed patients to be shifted directly from the CT scanner to the treatment machine without changing the patient s position. The CT scans were obtained immediately before treatment twice weekly, usually on Monday and Thursday. The CT on rails (General Electric Medical Systems) was a GE High Speed X/I model. The average number of scans obtained during a 5-week course of IMRT was 9.6 (range, 8 11). All scans encompassed the entire treatment field using 3-mm slices. The bladder and rectum were contoured on each slice. Vaginal markers were also identified and marked on each scan. Statistical analysis Bladder and rectal volumes were calculated from each pretreatment and intratreatment scan using the initial full-bladder planning CT scan. To determine changes in the positions of platinum fiducial markers, all CT scans were imported to the Syntegra software suite (Philips Radiation Oncology Systems). Each intratreatment scan was fused with the initial full-bladder scan, using the CT CT crosscorrelation algorithm, which seeks to minimize the mean-square gray-value difference between two images, aligning scans to match bony structures. The quality of each image registration was reviewed and verified by a physician or physicist. Using these fusions, the absolute differences between the intratreatment seed positions and the initial full-bladder seed positions were measured in the x, y, and z directions. These differences were then averaged for each of the 16 patients. We also evaluated the maximum shifts observed in each case (which allowed us to determine the probability that any treatment during a course of IMRT would fail to cover the target because of daily variation). RESULTS The patient and tumor characteristics of the 24 patients in the trial are summarized in Table 1. Bladder volumes The mean full and empty bladder volumes measured from the initial planning CT scans were 480 ml (range, ) and 155 ml (range, ), respectively. However, despite written and oral instruction regarding the importance

3 258 I. J. Radiation Oncology d Biology d Physics Volume 82, Number 1, 2012 Table 1. Patient and tumor characteristics Characteristic Number (%) of patients Race/ethnicity White 19 (79) Black 2 (8) Hispanic 2 (8) Asian 1 (5) Age (y) #40 3 (12) (29) (37) $60 5 (21) Primary diagnosis Endometrial cancer 14 (58) Cervical cancer 10 (42) Type of surgery TAH and BSO 2 (8) TAH and BSO and lymph node sampling and 12 (50) staging Vaginal hysterectomy and staging 2 (8) Type 2 radical hysterectomy 2 (8) Type 3 radical hysterectomy 6 (25) Weight (kg) #60 3 (12) (42) (21) > (25) Current chemotherapy No 14 (58) Yes 10 (42) Abbreviations: BSO = bilateral salpingo-oophorectomy; TAH = total abdominal hysterectomy. and method of pretreatment fluid intake, most patients had wide variations in their bladder volumes during treatment. Figure 1 shows the initial full and empty bladder volumes and subsequent intratreatment bladder volumes for each patient. In most cases, large variations were seen in ontreatment bladder volumes despite careful counseling of patients before and during treatment. Seven of the 24 patients never again achieved a bladder volume as high as that measured at the initial full-bladder simulation. Despite several reminders, 1 patient (Patient 15) had an empty bladder during all of her treatments, with volumes consistently less than on her initial empty bladder scan. By contrast, several patients achieved intratreatment bladder volumes that were significantly greater than the volumes at the initial full-bladder simulation. The bladder volume on each intratreatment scan was compared with the bladder volume on the initial full-bladder scan, and the mean difference was calculated for each patient. The median mean difference was cc (range, ). The median difference between the maximum and minimum bladder volumes during treatment was 247 cc (range, ). Rectal volumes Patients were not given instructions concerning rectal evacuation. Figure 2 shows the rectal volumes measured from initial simulation and intratreatment CT scans. Although some patients had fairly consistent rectal volumes before and during treatment, others had marked variations in their rectal volumes. In some patients (e.g., Patients 6 and 9) the rectal volume at the time of simulation was substantially greater than volumes measured from scans taken during treatment. The median rectal volume during treatment was 104 cc (range, ). The median difference between the maximum and minimum rectal volumes during treatment was 56.2 cc (range, ). No correlation was found between rectal volume and treatment week. Variations in vaginal position Changes in rectal and bladder volumes contributed to variations in the position of the vaginal apex in the 16 patients who had platinum fiducial markers present throughout treatment (Table 2). In the anterior posterior direction, the mean shifts ranged from 0.3 to 1.71 cm (median, 0.64 cm). In the superior inferior direction, the mean shifts ranged from (cc) the Bladder Volume of the Rectum (cc) Volume of Patients Patients Fig. 1. Bladder volumes during treatment for each of the 24 patients in the study. Horizontal lines represent the bladder volumes on the initial full-bladder and empty-bladder computed tomography scans used for treatment planning; open circles indicate the bladder volumes calculated from intratreatment scans. Fig. 2. Rectal volumes during treatment for each of the 24 patients in the study. Horizontal lines represent the rectal volumes on the initial full-bladder computed tomography scan used for treatment planning; open circles indicate the rectal volumes calculated from intratreatment scans.

4 Vaginal movement during pelvic IMRT d A. JHINGRAN et al. 259 Table 2. Absolute movement of vaginal markers in centimeters Mean Maximum Patient no. Anterior posterior Superior inferior Right left Anterior posterior Superior inferior Right left 1* * * * * * * Mean Median Minimum Maximum Standard deviation * Two seeds were visible on the computed tomography scan, and the number in the chart represents the mean of the movements of the two seeds. to 1.45 cm (median, 0.62 cm). The smallest mean shifts were seen in the right left direction, where they ranged from 0 to 0.52 cm (median, 0.23 cm). The maximum shift in the anterior posterior direction ranged from 0.8 to 2.79 cm (median, 1.46 cm). In the superior inferior direction, the maximum shift ranged from 0.6 to 2.1 cm (median, 1.2 cm). In the right left direction, the maximum shift ranged from 0 to 0.9 cm median, 0.59 cm). The direction and degree of vaginal motion varied from patient to patient. One patient (Patient 7) had a 2.79-cm shift in the anterior posterior direction with very little change in the right left and superior inferior directions. Another patient (Patient 15) had large movements in the superior inferior direction but not much movement in the other two directions. Yet another patient (Patient 14) had very little change in any direction despite changes in bladder filling (Fig. 1). Figure 3 shows the relationships between bladder or rectal volumes and shifts in the positions of vaginal apex markers. In six cases, variations in bladder volume were significantly correlated with posterior anterior or inferior superior shifts in vaginal seed positions. Variations in rectal volume were also found to correlate significantly with seed position in six cases; the direction of shifts produced by variations in rectal filling seemed to be predominantly in the anterior posterior direction. In 11 of 16 cases, significant shifts in the location of the vaginal apex were correlated with variations in the volume of the bladder or rectum. DISCUSSION Posthysterectomy pelvic radiation therapy improves local control and in some cases survival in patients with high-risk cervical or endometrial carcinoma. However, concerns about acute and chronic side effects have led clinicians to limit radiation dose and to search for ways of delivering effective treatment with less risk. Highly conformal inverse-planned forms of radiation therapy, such as IMRT, have recently made it possible to reduce the dose delivered collaterally to critical structures during postoperative pelvic radiation therapy (1 3), and there is evidence that both acute and chronic bowel side effects are less common when the pelvic nodes and vagina are treated with IMRT rather than conventional conformal four-field techniques (4, 5). Although several reports have described the treatment planning advantages of IMRT, we believe that our study is the first to systematically study changes in the location of the vaginal apex and variations in bladder and rectal volume during a course of postoperative IMRT. Our results demonstrate that variations in the position of the vaginal target volume are often large, suggesting that internal organ motion must be considered during radiation therapy planning and delivery (Figure 4). To avoid underdosage of the target, some method must be used to define a planning target volume that accounts for all uncertainties in the position of the CTV during a course of radiation treatment. The potential for underdosage due to internal variations in target position can theoretically be overcome in several ways. A very large margin may be added to the CTV that would account for even the largest variations seen in our study. In the past, we have used this approach, typically simulating patients with a very full bladder (which tends to maximize the posterior and superior extent of the target) and then adding a margin of several centimeters anteriorly to accommodate potential variations in bladder filling

5 260 I. J. Radiation Oncology d Biology d Physics Volume 82, Number 1, 2012 Fig. 3. Correlations between bladder or rectal volumes and the posterior anterior and inferior superior positions of markers inserted in the vaginal apices of 16 patients. Numbers on the y axis represent the organ volumes in cubic centimeters. The correlation coefficient (r) is indicated for each comparison; significant correlation coefficients are shown in red and marked with an asterisk (*). The significant correlation coefficient corresponds to a p value less than or equal to 0.05.

6 Vaginal movement during pelvic IMRT d A. JHINGRAN et al. 261 Fig. 4. Fused bladder, rectal, and vaginal contours for 2 patients in the study. Left, empty and full bladder contours are shown in solid and hatched red, respectively. The cephalad extent of the full bladder vaginal contour (yellow) extends approximately 2 cm superior to that of the empty bladder contour (green). Right, dramatic anterior deviation of the vagina with a change in rectal filling; the yellow vaginal contour corresponds to the relatively empty (solid blue) rectal contour, and the anteriorly displaced green vaginal contour corresponds to the expanded blue outline of the rectum on the second scan. In both cases, the superior extent of the vaginal apex was determined by the location of fiducial markers. and, if the rectal volume was large at the time of simulation, also a relatively large posterior margin. This approach reduced the risk of underdosage but potentially treated a large volume of tissue unnecessarily and is probably unacceptable when high-risk features necessitate a dose of more than 45 to 50 Gy to the vaginal target. In our view, a better approach is to individualize the ITV by fusing planning CT scans taken with full and empty bladder to estimate the potential range of positions during treatment. This accounts for much of the variation in the position of the vaginal apex, although the potential influence of rectal variations still must be considered in treatment planning. As our study shows, with this method, some patients will require relatively small margins to account for internal organ motion. Although the variation in target location could probably be diminished by treating patients with an empty bladder, this would result in significantly greater volumes of bladder and small bowel treated to high doses. The small bowel is probably the most important dose-limiting structure during pelvic irradiation, and the volume of the small bowel exposed to radiation has been linked to both acute and late bowel injury (6, 7). Although a variety of surgical (8, 9) and nonsurgical techniques have been used in attempts to reduce the volume of small bowel in the treatment field, bladder distention is probably the simplest technique. Several investigators have reported that bladder distention during radiation therapy significantly reduced the volume of bowel in the field (7, 10, 11). It was our hope, at the beginning of our study, that by giving patients careful directions for filling their bladder daily, we could achieve greater consistency in internal target location. However, as our study shows, even in the context of this prospective trial with careful instructions, the consistency of bladder filling was very poor. Our method of determining an ITV, though useful, should still be applied with caution. Several of our patients had one or more intratreatment bladder volumes that were greater than the bladder volume on the initial full-bladder scan, and other patients had evidence of more complete emptying during treatment than was observed at simulation, perhaps in part due to the gradual resolution of postsurgical changes and resolution of transient postoperative bladder atony. There is some evidence that bladder volumes tend to decrease, resulting in concomitant increases in small bowel exposure, over the course of radiation therapy (12, 13). The authors of one study (13) hypothesized that patients had greater difficulty maintaining a full bladder as they began to experience symptoms of bladder, small bowel, or rectal irritation or that, alternatively, patients simply grew less attentive to weekly reminders. However, we found no correlation in this study between time since the start of the radiation therapy course and bladder or rectal volume. One approach we have used to decrease the uncertainty about target location is to fill the bladder with a fixed volume of saline (via a Foley catheter) during simulation and

7 262 I. J. Radiation Oncology d Biology d Physics Volume 82, Number 1, 2012 immediately before treatment. This is a valuable method that can be used during a short course of boost treatment for patients who have particularly high-risk local features, but it is probably not practical for an entire course of pelvic IMRT. For patients who have marked shifts in the position of the vaginal apex between full and empty bladder scans, on-treatment ultrasound or CT can be used to confirm that bladder volumes are within 5% of the initial treatment planning bladder volume. Most patients seem to tolerate this well. Another approach that has been suggested is to use image guidance to make daily adjustments in field position based on the location of fiducial markers implanted in the vagina (14). This may be useful in selected cases if the target is limited to the vagina. However, when the vagina is treated at the same time as the pelvic nodes, it is important to remember that internal variations in vaginal position are largely independent of the nodal target volume. Any attempt to use isocenter adjustments to compensate for the large variations in vaginal position seen in some of our patients would cause variations of similar magnitude in the relative positions of the isocenter and nodal CTV; this would invariably cause underdosage of the target or require unacceptable nodal planning target volumes. Most of these maneuvers are meant to address the impact of variations in bladder volume. However, our study demonstrates that variations in rectal filling can also have a major impact on vaginal position, particularly in the anterior posterior direction. Men who are receiving radiation therapy for prostate cancer are sometimes given a Fleets enema before treatment, minimizing variations in the prostate position caused by rectal filling. This approach is probably not practical for gynecology patients, who, because of their larger treatment volumes, are often troubled with diarrhea, hemorrhoids, and other local problems that add to the discomfort of frequent enemas. However, we have found that it is prudent to evaluate rectal volume at the time of simulation. If the rectal ampulla is expanded by feces or gas, we ask the patient to attempt to evacuate, or we place a rectal tube to eliminate the gas. If these methods are unsuccessful, we increase the posterior vaginal contour by 5 to 10 mm to account for possible variations during treatment. Fiducial makers can also be followed during treatment to verify that variations in rectal volume are not causing the target to move out of the treatment area. From this study and from a similar study that we have completed in patients with intact cervical cancer (15), we have learned to have much greater respect for the challenges posed by internal organ motion in patients treated with radiation for gynecologic malignancy. Combinations of the above approaches can diminish the impact of variations in organ position on the adequacy of treatment and on the exposure of critical structures to radiation. However, the impact of changes in organ position cannot be ignored without compromising the likelihood of successful treatment outcome. REFERENCES 1. Ahamad A, D Souza W, Salehpour M, et al. Intensity-modulated radiation therapy after hysterectomy: Comparison with conventional treatment and sensitivity of the normal-tissuesparing effect to margin size. Int J Radiat Oncol Biol Phys 2005;62: Heron DE, Gerszten K, Selvaraj RN, et al. Conventional 3D conformal versus intensity-modulated radiotherapy for the adjuvant treatment of gynecologic malignancies: A comparative dosimetric study of dose-volume histograms small star, filled. Gynecol Oncol 2003;91: Roeske JC, Lujan A, Rotmensch J, et al. Intensity-modulated whole pelvic radiation therapy in patients with gynecologic malignancies. IntJRadiatOncolBiolPhys2000;48: Mundt AJ, Lujan A, Rotmensch J, et al. Intensity-modulated whole pelvic radiotherapy in women with gynecologic malignancies. Int J Radiat Oncol Biol Phys 2002;52: Mundt AJ, Mell LK, Roeske JC. Preliminary analysis of chronic gastrointestinal toxicity in gynecology patients treated with intensity-modulated whole pelvic radiation therapy. Int J Radiat Oncol Biol Phys 2003;56: Gallagher MJ, Brereton HD, Rostock RA, et al. A prospective study of treatment techniques to minimize the volume of pelvic small bowel with reduction of acute and late effects associated with pelvic irradiation. Int J Radiat Oncol Biol Phys 1986;12: Miller RC, Martenson JA, Sargent DJ, et al. Acute treatmentrelated diarrhea during postoperative adjuvant therapy for high-risk rectal carcinoma. Int J Radiat Oncol Biol Phys 1998;41: Herbert SH, Solin LJ, Hoffman JP, et al. Volumetric analysis of small bowel displacement from radiation portals with the use of a pelvic tissue expander. Int J Radiat Oncol Biol Phys 1993;25: Rodier JF, Janser JC, Rodier D, et al. Prevention of radiation enteritis by an absorbable polyglycolic acid mesh sling: A 60-case multicentric study. Cancer 1991;68: Georg P, Georg D, Hillbrand M, et al. Factors influencing bowel sparing in intensity modulated whole pelvic radiotherapy for gynaecological malignancies. Radiother Oncol 2006; 80: Han Y, Shin EH, Huh SJ, et al. Interfractional dose variation during intensity-modulated radiation therapy for cervical cancer assessed by weekly CT evaluation. Int J Radiat Oncol Biol Phys 2006;65: Brierley JD, Cummings BJ, Wong CS, et al. The variation of small bowel volume within the pelvis before and during adjuvant radiation for rectal cancer. Radiother Oncol 1994;31: Nuyttens JJ, Robertson JM, Yan D, et al. The position and volume of the small bowel during adjuvant radiation therapy for rectal cancer. Int J Radiat Oncol Biol Phys 2001;51: Yamamoto R, Yonesaka A, Nishioka S, et al. High dose threedimensional conformal boost (3DCB) using an orthogonal diagnostic X-ray set-up for patients with gynecological malignancy: A new application of real-time tumor-tracking system. Radiother Oncol 2004;73: Beadle BM, Jhingran A, Salehpour M, et al. Cervix regression and motion during the course of external beam chemoradiation for cervical cancer. Int J Radiat Oncol Biol Phys 2009;73: