|Year : 2019 | Volume
| Issue : 8 | Page : 69-75
Interfraction prostate displacement during image-guided radiotherapy using intraprostatic fiducial markers and a cone-beam computed tomography system: A volumetric off-line analysis in relation to the variations of rectal and bladder volumes
Gianluca Ingrosso1, Roberto Miceli1, Elisabetta Ponti1, Andrea Lancia1, Daniela di Cristino1, Francesco de Pasquale2, Pierluigi Bove3, Riccardo Santoni1
1 Department of Diagnostic Imaging, Molecular Imaging, Interventional Radiology and Radiotherapy, Tor Vergata General Hospital, Rome, Italy
2 Department of Radiology, Santa Lucia Foundation, Rome, Italy
3 Department of Surgery, Urology Unit, Tor Vergata General Hospital, Rome, Italy
|Date of Web Publication||22-Mar-2019|
Dr. Gianluca Ingrosso
Tor Vergata General Hospital, 00133 Rome
Source of Support: None, Conflict of Interest: None
Purpose: Prostate motion during the radiotherapy course is an important issue. This study investigated the inter-fraction prostate motion in controlled rectal filling condition.
Methods: 10 prostate cancer patients underwent image-guided radiotherapy (IGRT) using a cone-beam computed tomography (CBCT) system, after the insertion of fiducial markers (FMs). The planning CT was the reference CT (CTref) used to estimate the reference intermarker distances, and CBCTs were used for off-line comparison with CTref. We evaluated the influence of rectal and bladder volume on prostate shifts. We calculated the required planning target volume (PTV) margins in this patient population.
Results: 120 CBCTs were analyzed. Mean prostate displacements (± SD) along the 3 axes (x, y, z) averaged over the 10 patients, were: 0.90 ± 0.84 mm in x, 0.00 ± 2.07 mm in y, -0.80 ± 1.28 mm in z. There is a statistically significant anti-correlation between prostate displacements and: bladder volume variations (P < 0.001) in the y-axis, and rectal volume variations (P < 0.05) in the z-axis. PTV margins obtained for the directions x, y and z are respectively 2.5, 5.6 and 3.9 mm.
Conclusion: IGRT in reproducible empty rectum condition allow a high reduction of daily treatment uncertainties.
Keywords: Image-guided radiotherapy, inter-fraction displacement, prostate cancer, volumetric off-line analysis
|How to cite this article:|
Ingrosso G, Miceli R, Ponti E, Lancia A, di Cristino D, de Pasquale F, Bove P, Santoni R. Interfraction prostate displacement during image-guided radiotherapy using intraprostatic fiducial markers and a cone-beam computed tomography system: A volumetric off-line analysis in relation to the variations of rectal and bladder volumes. J Can Res Ther 2019;15, Suppl S1:69-75
|How to cite this URL:|
Ingrosso G, Miceli R, Ponti E, Lancia A, di Cristino D, de Pasquale F, Bove P, Santoni R. Interfraction prostate displacement during image-guided radiotherapy using intraprostatic fiducial markers and a cone-beam computed tomography system: A volumetric off-line analysis in relation to the variations of rectal and bladder volumes. J Can Res Ther [serial online] 2019 [cited 2019 Nov 15];15:69-75. Available from: http://www.cancerjournal.net/text.asp?2019/15/8/69/228997
| > Introduction|| |
Daily patient setup is an important step in prostate cancer radiotherapy. Pelvic organs mobility and their volume variations (i.e., rectum and bladder) lead to a discrepancy between the snapshot of the planning computed tomography (CT) and the actual pelvic volumes during the treatment course. In particular, it is observed that displacements resulting from prostate motion are more significant than setup errors, and these uncertainties may have important clinical consequences. For instance, two randomized studies, demonstrated that a distended rectum at the planning CT significantly reduced local control, because of a systematic difference between planned and actual position of the prostate during the treatment course. To overcome prostate position uncertainties, on-line and off-line imaging devices have been developed since now, and are currently used to avoid the target geographic miss, minimizing normal tissue irradiation at the same time. A number of studies evaluating prostate motion during the treatment course using multiple CT scans or implanted markers,,, and procedures to keep the rectum empty have been described.,
In this study, we evaluated in ten patients the interfraction prostate gland motion, in empty rectum and full bladder condition during the treatment course, using intraprostatic fiducial markers detected on planning CT images and matched offline with cone-beam CT (CBCT) images acquired for image guidance. We also calculated the required planning target volume (PTV) margins using measured shits in this patient population.
| > Materials and Methods|| |
From April 2010 to March 2011, ten consecutive patients diagnosed with low risk localized prostate cancer underwent three-dimensional (3D) conformal image-guided radiotherapy (IGRT) using CBCT, after the transperineal insertion of intraprostatic Visicoil under trans-rectal ultrasound guidance. Visicoils features and implantation procedure have been previously described. Briefly, a total of 29 Visicoil™ (diameter 0.75 mm and length 10 mm) were implanted (two into each lobe of the prostate toward the base and one toward the apex), three in each patient except one, who had only two FMs implanted because of a previous transurethral prostate resection. All patients underwent CT (2.5 slice thickness) and MRI under radiotherapy planning conditions in supine position; bowel and bladder preparation was prescribed to have an empty rectum and a full bladder during the CT scan and the treatment course. For each patient, the clinical target volume (CTV), consisting of the prostate gland, and organs at risks were outlined. In particular, rectum and bladder were defined as solid organs (what is mean by solid organ); the rectum was contoured from rectosigmoid junction to the lowest level of the ischial tuberosities, and the bladder was contoured in its entirety. PTVs were generated by an asymmetric expansion of CTVs (5 mm at the posterior margin and 6 mm in all other directions). Conformal treatment plans 3D conformal radiation therapy (CRT) was obtained on Pinnacle3 version 8.0 m (Philips Medical System, Andover, MA, USA). A total dose of 76 Gy (2 Gy per fraction) was prescribed to the PTV, and radiotherapy treatment was delivered using Elekta Synergy® S linear accelerator equipped with a kV cone-beam CT system., Intraprostatic FMs detected on planning CT images were matched online with those on CBCT images for image guidance, acquired in the first 5 days of treatment and then once a week until the end of the treatment.
For the purpose of this study, the planning CT was the reference CT (CTref) used to estimate the reference intermarker distances, and CBCTs were used for offline comparison with CTref. Reconstructed images of CBCTs (410 × 164 × 410 voxel matrix with voxel dimension 1 mm × 1 mm × 1 mm) were exported to Pinnacle3 8.0 m treatment planning System (TPS) through Digital Imaging and Communications in Medicine and imported as secondary fusion datasets into the TPS image fusion module, named Syntegra. For every patient, after the registration procedure between CTref and each CBCT using mutual information algorithm, the correspondence of the bony anatomy was checked. On CTref and on each imported CBCT, the same physician contoured every FM, together with rectum and bladder. At the center of mass (CM) of each contoured FM, we created a point of interest (POICM) using the automatic positioning “centroid” tool, implemented on the TPS. The relative coordinates (x: Right [−] – left [+], y: Posterior [−] – anterior [+], and z: Inferior [−]-superior [+]) of the POICM were recorded. To test the automatic positioning of the POICM, the procedure was repeated few times, comparing every time the 3 coordinates (x, y, z) of the POICM: No variations in the position were determined.
In every patient, we computed the differences between the coordinates (x, y, and z) of the POICM for each corresponding FM of the two studies: CTref and CBCT. Average values, the representative of prostate gland movements along the 3 axes, were obtained from the shifts of the three markers (two for one patient).
To identify shifts statistically larger than zero, after checking the normality of the acquired data, we performed a t-test (P < 0.05). Now, to check if there is a correlation between the prostate shift and rectal and bladder volumes changes compared with volumes registered on CTref, we computed a Pearson correlation coefficient between these variables. Eventually, to evaluate the influence of rectal and bladder volume on the collected prostate shifts, we performed a linear regression analysis adopting the former variables as independent and the latter one as a dependent.
Finally, requiring that 95% of the patients achieve a minimum dose of 95% from prescribed dose to the CTV, the CTV-to-PTV margin calculation was performed applying the van Herk margin formula, written as:
1.96 Σ+0.7 σ
Where Σ is the standard deviation (SD) of all systematic errors, σ is the SD of random errors (calculated as the root mean square of the quadratic sum of the SDs of all patients).
| > Results|| |
A total of 120 CBCTs were analyzed, with a mean of 12 CBCTs for the patient (range 11–15 CBCTs for each patient) acquired in the first 5 days of treatment, and then, once a week until the end of the treatment. The mean of prostate displacements (± SD) along the three main axes (x, y, and z) averaged over the ten patients, and evaluated by the shifts of the gold coils, were: 0.90 ± 0.84 mm in x, 0.00 ± 2.07 mm in y, and −0.80 ± 1.28 mm in z whereas absolute shifts were: 1.20 ± 0.65 mm in x, 2.10 ± 0.71 mm in y, and 1.50 ± 0.80 mm in z.
As shown in [Figure 1], we reported the mean and standard errors and the frequency distribution of prostate shifts, relatively to all patients along the three axes. We obtained significant shifts in the left-right (L-R) direction (x) in four patients (range + 1/+2.7 mm), in the anterior-posterior (A-P) direction (y) in five patients (range −2.8/+2.7 mm), and in the superior-inferior (S-I) direction (z) in four patients (range −3.5/+1 mm). Mean prostate displacements in x-axis are always in the left direction, in y-axis, they are quite equally distributed, and in z-axis, they are mainly in the caudal direction [Table 1]. Demonstrated the stable position of FMs during the whole radiotherapy treatment, shift values obtained along the 3 axes represent the displacement of the prostate due primarily to rectal and bladder filling variations. To control for this effect, for every patient, we reported volumes (cm3) of rectum and bladder of CTref and CBCTs (mean values), and we computed the relative variations between the bladder and rectal volumes obtained from CBCTs and those from CTref[Table 2]. It is interesting to note that, although patients received instructions to void their bladder and drink 250 ml of water 1 h before planning CT and treatment sessions, the bladder undergoes a greater change in volume compared to the rectum [Figure 2]. In fact, we recorded for the total cohort a range of relative rectal volume variations from −14% to 39.8%, and a range of relative bladder volume variations from −46% to 127.2%, respect to CTref.
|Figure 1: Average shifts and frequency distribution of prostate shifts for each patient in x (L-R), y (A-P), z (S-I) directions, respectively|
Click here to view
|Table 1: Average motion of the prostate in 10 patients (120 cone-beam computed tomography)|
Click here to view
|Table 2: Reference computed tomography, and cone-beam computed tomography's rectal and bladder volumes (mean value, range, and percentage relative standard deviation) of the 10 patients (120 cone-beam computed tomography's).|
Click here to view
|Figure 2: Rectum and bladder volumes variations during the treatment. For every patient, the volume at reference computed tomography was compared with the mean volume of cone-beam computed tomography|
Click here to view
To assess a possible relationship between prostate displacements along the 3 axes and changes in rectal and bladder volumes during the treatment course, we computed the Pearson correlation between these variables. We obtained that in the x-axis there is a trend toward significance in the correlation between the shift of the prostate and rectal volume variations (P < 0.08): The increase of the rectal volume tends to shift the prostate in the left direction. In the y-axis, there is a statistically significant correlation between prostate displacements and bladder volume variations (r = −0.58; P < 0.001): During the radiation treatment course, the bladder-volume increase pushes the prostate gland in the posterior direction, while the bladder-volume decrease moves the prostate in the anterior direction [Figure 3]. In the z-axis, there is no statistically significant correlation between prostate displacements and rectal and bladder volume variations. Eventually, we performed a linear regression analysis to test the possible linear model corresponding to the shift along y-axis as a function of independent variable (bladder-volume). In particular, [Figure 3] shows the scatterplot of the y-shifts versus the normalized volume of the bladder (dots). The regression analysis (solid line) revealed a significant linear relationship between these two variables corresponding to R2 = 0.34. Notably, in this analysis, only significant values of these two variables, previously extracted, were included. This is due to the fact that from a clinical point of view, our experimental question is whether there is a linear relationship between shifts and volumes when they are significantly different from zero. As matter of fact, from a clinical standpoint, nonsignificant shifts or volume variations can be considered negligible. Moreover, when the values are very small, these estimates can be very different from the range in which they are significant, and this larger error might lead to apparent complex regimes, for example, a combination of linear and nonlinear effects. However, since from a mathematical point of view to remove such nonsignificant values might inflate the obtained correlation and regression coefficients, we have run the same analysis on all the acquired data, i.e., including nonsignificant values. We obtained a still significant statistically correlation (r = −0.187; P < 0.001), and a linear relationship between the data (R2 = 0.035) although at a lower strength (as expected), due to the presence of large proportion (≈40%) of values close to zero. This is an important control, which validates our previous result.
|Figure 3: Scatterplots of prostate shifts in the y (A-P) direction relative to normalized bladder volume variations, restricted to significant values (above), and including all the acquired data (below). Dotted lines represent 95% confidence intervals for the linear fit|
Click here to view
Finally, we estimated the CTV-to-PTV margins through the formula of van Herk. The values obtained for the directions x, y, and z are respectively 2.5, 5.6, and 3.9 mm.
| > Discussion|| |
To the best of our knowledge, this is the first volumetric analysis using on-board CBCT images combined with fiducial markers to evaluate prostate motion in patients with stable empty rectum (absolute mean rectal volume variation of 14.8%, for the total cohort) during the treatment course.
Patients were treated with conformal radiotherapy using a conventional fractionation (38 × 2 Gy). Although new techniques such as Intensity-modulated radiation therapy and volumetric modulated arc therapy are currently available in the treatment of localized prostate cancer, 3DCRT is still considered the standard of care. IGRT with cone-beam CTs was performed every day during the first 5 days of treatment, and then once a week. We used this IGRT schedule to perform an adaptive radiotherapy based on the check of the first 5 days. In the current study, we collected volumetric CBCT data of these patients to carry out a retrospective off-line analysis. Our off-line evaluation showed that in empty rectum and full bladder conditions prostate displacement during the treatment course is very limited, with a maximum average prostate motion in the 3 axes equal to: 2.7 mm (only patient 9) in L-R direction, 2.8 mm in A-P direction, and 3.5 mm (only patient 1) in S-I direction. We found a trend toward significance (P < 0.08) between rectal volume variations and prostate displacement in the x-axis (L-R direction). In the y-axis (A-P direction), there was a statistically significant correlation (P < 0.001) between prostate displacements and bladder volume variations. Finally, in the z-axis (S-I direction) there were no statistically significant correlations between prostate displacements and rectal and bladder volume variations. The physiological behavior of rectum and bladder may help to explain our results. With regard to the rectum, though not significant in our analysis, there might be a correlation between the displacement of the prostate in the left direction and rectal filling. This result could be justified by the anatomical and functional features of the rectum itself, in particular to the left-anterior curvature at the rectosigmoid junction. In fact, several studies demonstrated,, that the rectum is anatomically more fixed at its caudal and dorsal end than at the cranial side. Hoogeman et al. quantified the shape and position changes of the rectum during the treatment course and between planning and treatment, using data of repeated CT scans in 19 prostate cancer patients. They demonstrated that rectum displacements were localized to its upper anterior, left, and right side, whereas the shape variations were small in the lower rectal region. Muren et al., evaluating the planning organ at risk volume (PRV) margins for the rectum, evidenced that the rectal motion at the upper superior part is larger to the left than to the right due to the left anterior curvature of the rectum at the rectosigmoid junction. Hence, the rectal volume increase in its upper part might shift the prostate in the left and inferior directions. In our analysis, these shifts are very small (absolute mean value in L-R 1.2 ± 0.65 mm, in S-I 1.5 ± 0.8 mm), and this is due to the small rectal volumes registered in CTref and CBCTs [Table 2].
With regard to the bladder, the correlation between its volume increase and the shift of the prostate in the posterior direction might be explained by the distension of the bladder in the presence of an empty rectum. As demonstrated by Lotz et al., rectal filling may have an influence in the bladder position, so in empty rectum conditions, the posterior expansion of the bladder is not hampered.
Our methodology and results are quite similar to the ones of Zellars et al. They analyzed 24 patients who underwent a planning CT scan (in supine position), in full bladder condition, and a second CT after 4–5 weeks of radiotherapy. Comparing the CM of the prostate, they demonstrated that the prostate shifted along a diagonal axis extending from an anterior-superior position to a posterior-inferior position, and that bladder movement and volume change and upper rectum movement were independently associated with prostate motion. Interestingly, they demonstrated that the (whole) rectum had no discernable association with prostate position, whereas the reanalysis of data with the rectum split into superior and inferior halves revealed a significant relationship between the prostate displacement in the posterior-inferior direction and the upper rectum movement. As reported in [Table 3], many authors,,,, demonstrated that the increase of the bladder volume causes the posterior shift of the prostate, while only Zellars reported a relation between the upper rectum movement and the prostate displacement in the posterior-inferior direction. It is to say that the majority of these studies evaluated the relationship between rectal filling and prostate motion with a free rectum state or under nonphysiologic conditions (administration of rectal contrast). In our analysis, we found no relationship between the rectal volume changes and the shifts of the prostate, and we believe that this result is mainly due to the real and quite stable empty rectum condition in our series [Figure 2]. As reported in [Table 2], using only a dietary regimen we obtained a mean rectal volume on planning CT that ranged from 26.57 cm3 to 66.50 cm3 (mean total value: 37.38 cm3), compared to a mean rectal volume on CBCTs that ranged from 22.79 cm3 to 79.81 cm3 (mean total value: 39.53 cm3). A comment is needed for patient 7, who shows the highest rectal and bladder increase respect to CTref without significant prostate displacements. The analysis of CBCTs for this patient shows a parallel volume increase of rectum and bladder, in every set of images. In particular, the mean rectal volume increase (up to 40% of the CTref volume) causes a whole organ filling (not only in its upper part), minimizing the prostate shift in the caudal direction. On the other hand, the huge volume bladder increase (up to 127% of the CTref volume) balances the possible anterior prostate motion due to the whole rectum filling, probably determining the recorded minimal prostate displacements.
|Table 3: Comparison between studies correlating prostate displacement with bladder and rectum state|
Click here to view
We demonstrated that using a simple dietary regimen, it is possible to obtain a quite stable empty rectum condition, with an absolute mean rectal volume variation of 14.8% for the total cohort (range of relative variations, −14%–39.8%). In these rectal volume conditions during the treatment course, there is a strong correlation between prostate movements and bladder volume changes (absolute mean bladder volume variation of 51%; range of relative bladder volume variations, −46%–127.2%). Finally, from our analysis, the CTV-to-PTV margins needed for the adequate radiation treatment are 2.5, 5.6, and 3.9 mm, respectively in x, y, and z directions. Further evaluation on a larger patient cohort with a daily imaging evaluation is needed before applying these margins in our clinical practice.
The main limitation of our work is the small number of patients, requiring an evaluation on a larger patient cohort. Another limitation is that data about prostate displacements were not on a daily basis. It is to say that we analyzed a quite large number of volumetric data on a total of 120 CBCTs. We performed a real prostate interfraction displacement evaluation under radiotherapy conditions, in patients who were only invited to have a dietary regimen in combination with mild laxatives, and a full bladder for planning CT and treatment sessions. We used a very precise system to calculate the CM of the prostate with stable intraprostatic fiducial markers, and 2.5 mm thickness planning CTs and 1 mm thickness kV-CBCTs for image registration.
In reproducible empty rectum condition, with a relative rectal volume variation from −14% to 39.8% during the treatment course, prostate movements depend only on bladder volume changes. The use of image-guidance with an onboard kV-CBCT system may allow a high reduction of daily treatment uncertainties related to setup and organ motion. Using van Herk formula to our population, we obtained a very tight expansion margins. Further evaluation on a larger patient cohort with a daily imaging evaluation is needed before applying these margins in our clinical practice.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| > References|| |
Alasti H, Petric MP, Catton CN, Warde PR. 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;49:869-84.
de Crevoisier R, Tucker SL, Dong L, Mohan R, Cheung R, Cox JD, et al.
Increased risk of biochemical and local failure in patients with distended rectum on the planning CT for prostate cancer radiotherapy. Int J Radiat Oncol Biol Phys 2005;62:965-73.
Heemsbergen WD, Hoogeman MS, Witte MG, Peeters ST, Incrocci L, Lebesque JV, et al.
Increased risk of biochemical and clinical failure for prostate patients with a large rectum at radiotherapy planning: Results from the dutch trial of 68 GY versus 78 gy. Int J Radiat Oncol Biol Phys 2007;67:1418-24.
Dehnad H, Nederveen AJ, van der Heide UA, van Moorselaar RJ, Hofman P, Lagendijk JJ, et al.
Clinical feasibility study for the use of implanted gold seeds in the prostate as reliable positioning markers during megavoltage irradiation. Radiother Oncol 2003;67:295-302.
Litzenberg D, Dawson LA, Sandler H, Sanda MG, McShan DL, Ten Haken RK, et al.
Daily prostate targeting using implanted radiopaque markers. Int J Radiat Oncol Biol Phys 2002;52:699-703.
van der Heide UA, Kotte AN, Dehnad H, Hofman P, Lagenijk JJ, van Vulpen M, et al.
Analysis of fiducial marker-based position verification in the external beam radiotherapy of patients with prostate cancer. Radiother Oncol 2007;82:38-45.
Stasi M, Munoz F, Fiorino C, Pasquino M, Baiotto B, Marini P, et al.
Emptying the rectum before treatment delivery limits the variations of rectal dose – Volume parameters during 3DCRT of prostate cancer. Radiother Oncol 2006;80:363-70.
Nijkamp J, Pos FJ, Nuver TT, de Jong R, Remeijer P, Sonke JJ, et al.
Adaptive radiotherapy for prostate cancer using kilovoltage cone-beam computed tomography:First clinical results. Int J Radiat Oncol Biol Phys 2008;70:75-82.
Miceli R, Ingrosso G, Ponti E, di Cristino D, Lancia A, Bove PL, et al.
Evaluation of serrated intraprostatic gold coils positional stability using on-board cone beam computed tomography scans acquired throughout the radiation treatment course. Pract Radiat Oncol 2015;5:417-22.
Ingrosso G, Carosi A, Ponti E, Murgia A, di Cristino D, Barbarino R, et al.
Acute and late toxicity after three-dimensional conformal image-guided radiotherapy for localized prostate cancer. Cancer Invest 2014;32:526-32.
Falco MD, Fontanarosa D, Miceli R, Carosi A, Santoni R, D'Andrea M, et al.
Preliminary studies for a CBCT imaging protocol for offline organ motion analysis: Registration software validation and CTDI measurements. Med Dosim 2011;36:91-101.
van Herk M, Remeijer P, Rasch C, Lebesque JV. The probability of correct target dosage: Dose-population histograms for deriving treatment margins in radiotherapy. Int J Radiat Oncol Biol Phys 2000;47:1121-35.
Hoogeman MS, van Herk M, de Bois J, Muller-Timmermans P, Koper PC, Lebesque JV, et al.
Quantification of local rectal wall displacements by virtual rectum unfolding. Radiother Oncol 2004;70:21-30.
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;70:11-9.
Hoogeman MS, van Herk M, de Bois J, Lebesque JV. Strategies to reduce the systematic error due to tumor and rectum motion in radiotherapy of prostate cancer. Radiother Oncol 2005;74:177-85.
Lotz HT, Remeijer P, van Herk M, Lebesque JV, de Bois JA, Zijp LJ, et al.
Amodel to predict bladder shapes from changes in bladder and rectal filling. Med Phys 2004;31:1415-23.
Zellars RC, Roberson PL, Strawderman M, Zhang D, Sandler HM, Ten Haken RK, et al.
Prostate position late in the course of external beam therapy: Patterns and predictors. Int J Radiat Oncol Biol Phys 2000;47:655-60.
Ten Haken RK, Forman JD, Heimburger DK, Gerhardsson A, McShan DL, Perez-Tamayo C, et al.
Treatment planning issues related to prostate movement in response to differential filling of the rectum and bladder. Int J Radiat Oncol Biol Phys 1991;20:1317-24.
Schild SE, Casale HE, Bellefontaine LP. Movements of the prostate due to rectal and bladder distension: Implications for radiotherapy. Med Dosim 1993;18:13-5.
van Herk M, Bruce A, Kroes AP, Shouman T, Touw A, Lebesque JV, et al.
Quantification of organ motion during conformal radiotherapy of the prostate by three dimensional image registration. Int J Radiat Oncol Biol Phys 1995;33:1311-20.
Roeske JC, Forman JD, Mesina CF, He T, Pelizzari CA, Fontenla E, et al.
Evaluation of changes in the size and location of the prostate, seminal vesicles, bladder, and rectum during a course of external beam radiation therapy. Int J Radiat Oncol Biol Phys 1995;33:1321-9.
Zelefsky MJ, Crean D, Mageras GS, Lyass O, Happersett L, Ling CC, et al.
Quantification and predictors of prostate position variability in 50 patients evaluated with multiple CT scans during conformal radiotherapy. Radiother Oncol 1999;50:225-34.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3]