Intrafraction prostate motion during IMRT for prostate cancer

212

I. J. Radiation Oncology

● Biology ● Physics

Volume 51, Number 3, Supplement 1, 2001

cranio-caudal direction were observed. The results showed, that the seminal vesicles --being responsible for the dorsal field contour-- are more influenced by the bladder filling than the prostate gland being responsible for the ventral field border. The evaluated movements of the prostate and the seminal vesicles have to be taken into account when defining safty margins for radiotherapy treatment. This is of additional importance when it is not possible to assure the filling status of the bladder for each treatment fraction.

1071

Dancing Prostates: Measurement of Intra-Fractional Prostate Motion using MRI

1

D. Mah , G. Freedman1, B. Milestone2, B. Movsas1, R. Mitra1, E. Horwitz1, G.E. Hanks1 1 Radiation Oncology, Fox Chase Cancer Center, Philadelphia, PA, 2Diagnostic Imaging, Fox Chase Cancer Center, Philadelphia, PA Purpose: To quantify the three-dimensional intra-fractional prostate motion over typical treatment time intervals with cine MRI studies. Materials and Methods: Patients were scanned supine in an MRI with a flat tabletop insert and alpha cradle. Two distinct sets of measurements were made. First, to prospectively determine the motion of the prostate, 7 patients were scanned every 12 seconds for 6 minutes. A 1 cm thick slice through the widest aspect of the prostate was obtained in the axial and the sagittal planes. Prostate motion was measured by contouring the prostate on each slice and calculating its center of mass motion over time. To investigate the effect of respiratory motion, the same image slice was acquired every 1.3 s for 30 slices. Additionally, to determine how often significant prostate motions occurred, two MRI volume sets taken 7-10 minutes apart were retrospectively compared. The MRI images were fused according to their bony landmarks using a chamfer-matching program. The prostate was contoured and its position on the first volume set was compared to the second volume set for 23 patients. Results: No significant respiratory motion was detected. Cine loops of the prostate motion indicated that most of the motion was near the base, with the prostate appearing to precess about the apex due to peristaltic motion of the rectum. The standard deviation of the prostate motion from the cine images was 2.0, 1.8 and 1.1 mm in the AP, SI and ML directions respectively (range 0 to 9 mm). Significant motions of the prostate (⬎5 mm) occurred in 2 of the 23 patients examined retrospectively. Figure 1 shows a cross sectional image and Fig. 2 shows the typical motion of the prostate over a time course of 6 minutes. Conclusion: During radiation treatments, the prostate can move intra-fractionally with a standard deviation of ⬃ 2 mm due to peristalsis. Larger movements of 5 mm occur with a frequency of ⬍10%. This motion is less than typical setup errors and should not affect IMRT or conformal treatments of radiation therapy.

1072

Intrafraction Prostate Motion during IMRT for Prostate Cancer

E. Huang, L. Dong, A. Chandra, D.A. Kuban, I.I. Rosen, A. Pollack Radiation Oncology, UT-MD Anderson Cancer Center, Houston, TX Purpose: While the interfraction movement of the prostate has been previously studied through the use of fiducial markers, CT scans, and ultrasound-based systems, intrafraction movement during treatment is not well-documented. In this report the NOMOS (Sewickley, PA) BAT ultrasound system was used to measure intrafraction prostate motion during 100 IMRT treatments for prostate cancer. Materials and Methods: Ten men under treatment with IMRT for clinically localized prostate cancer, who had good pretreatment prostate images during the first few treatments, were selected for the study. All patients were treated in the supine position and received 75.6 Gy in 42 fractions. Pre- and post-treatment ultrasound alignment images were collected immediately before and after IMRT treatment on 10 treatment days, for a total of 200 procedures. Any BAT ultrasound misalignments of the prostate borders in relation to the planning CT-scan were recorded in 3 dimensions. The couch was then shifted to achieve alignment between the ultrasound and CT-plan outline images. The entire process of positioning, localization, and alignment was performed by experienced radiation therapists without physician guidance. One person (E.H.) reviewed every image and classified its quality and alignment on a 3-point scale: 1) near perfect image quality or alignment; 2) fair image quality or misalignment ⱕ5 mm (within the PTV); and 3) unacceptable image quality or misalignment ⬎5 mm (potential encroachment on the PTV). The magnitude of each shift during treatment was determined in 3 dimensions: lateral (RL), anterior-posterior (AP), and superior-inferior (SI). Each alignment was classified as either No Shift, Shift ⱕ2.5 mm, Shift ⬎2.5 to 5 mm, or Shift ⬎5 mm.

Proceedings of the 43rd Annual ASTRO Meeting

213

Results: All of the images were judged to be of acceptable quality and alignment. The dominant directions of intrafraction prostate motion were anteriorly and superiorly. The mean magnitudes of the shifts (⫾ SD) were 0.02 ⫾ 0.40 mm, 0.32 ⫾ 1.50 mm, and 0.13 ⫾1.02 mm in the RL, AP, and SI directions, respectively. The maximum range of motion occurred in the AP dimension, from 6.8 mm anteriorly to 3.5 mm posteriorly. The percentages of treatments with no post-treatment shifts were 77%, 64%, and 65% in the RL, AP, and SI directions, respectively. There were 15 shifts ⬎2.5 mm in 7 patients; 3 patients had 1 such shift, and 4 patients had 3 such shifts . Of these, 80% of the AP and 100% of the SI shifts were between ⬎2.5 - 5 mm. There were only two shifts, in two separate patients, that were ⬎5 mm (6.8 and 5.7 mm). The percentages of treatments during which prostate motion was judged to be safely within the PTV (shift ⱕ5 mm) were 100%, 98%, and 100% in the RL, AP, and SI directions, respectively. Conclusion: To our knowledge, this is the first ultrasound-based study describing prostate motion during radiotherapy. Our findings suggest that intrafraction movement occurs predominantly in the anterior and superior directions. In the vast majority of the treatments, the extent of prostate motion was safely within the PTV (shift ⱕ5 mm). Intrafraction prostate motion that potentially encroached on the PTV (shift ⬎ 5 mm) occurred in only 2% of alignments (AP dimension). Pretreatment prostate position by ultrasound adequately reflects prostate position during IMRT.

Direction LAT AP SI

1073

Mean⫾SD mm (Range)

No Shift

Shift ⱕ2.5 mm

Shift ⬎2.5 mm to 5 mm

Shift ⬎5 mm

0.02⫾0.40 (2.50 to -2.40) 0.32⫾1.50 (6.80 to -3.50) 0.13⫾1.02 (3.30 to -2.90)

77% 64% 65%

23% 26% 30%

0% 8% 5%

0% 2% 0%

3D Intra-Fractional Movement of Prostate Measured during Real-Time Tumor-Tracking Radiation Therapy (RTRT) in Supine and Prone Treatment Positions

K. Kitamura1, H. Shirato1, Y. Seppenwoolde2 1 Radiology, Hokkaido University School of Medicine, Sapporo, Japan, 2Radiotherapy, The Netherlands Cancer Institute, Amsterdam, Netherlands

Purpose: Intra-fractional movement of the prostate gland in the supine versus the prone position remains controversial. Real-time tumor-tracking radiation therapy (RTRT) with insertion of a fiducial marker may indicate which position allows for the best treatment. The purpose of this study was to quantify 3D movement of the prostate gland in the supine and prone positions and to analyze the frequency of movement. Materials and Methods: The RTRT was developed to recognize the 3D position of a 2 mm gold marker implanted in the prostate 30 times a second using 2 sets of fluoroscopic images. The linear accelerator is triggered to irradiate only when the marker is located planned coordinates relative to the isocenter. Four patients with prostate cancer treated with RTRT were the subjects of this study. The coordinates of the marker were recorded every 0.033 seconds during RTRT in the supine position for 2 minutes. The patient was then moved to the prone position, and the marker was tracked for 2 minutes to acquire data while in this position. Patients were not irradiated in the prone position. Measurement was performed 5 times for one patient; a total of 20 sets were analyzed. Fourier transform of the unfiltered data was performed for frequency analysis of movement. Results: The table shows the maximum, the mean, and the standard deviation (SD) of amplitude of 3D movements for each patient in the supine and prone positions. Amplitude of the 3D movement was 0.2-2.6 mm in the supine and 0.7-8.0 mm in the prone position. The amplitude while in the supine position was statistically smaller in all directions than those while in the prone position (p⬍0.0001). The amplitude in the cranio-caudal (C-C) and anterior-posterior (A-P) directions was larger than in the left-right direction in the prone position (p⬍0.02). No characteristic frequency of movement was detected in the supine position. Respiratory frequency (0.2-0.3 Hz) was detected for all patients in the movement in the C-C and A-P directions in the prone position. Frequency analysis suggested that prostate movement is affected by the respiratory cycle and is influenced by bowel movement in the prone position. Conclusion: This study confirmed that organ motion is less frequent in the supine position than in the prone position in the treatment of prostate cancer. RTRT would be useful to reduce uncertainty due to effects of the respiratory cycle, especially in the prone position. Max amplitude (mean⫾SD) in mm No.

Position

1

Supine Prone Supine Prone Supine Prone Supine Prone

2 3 4

Left-right 0.3 2.3 0.2 0.7 0.4 2.1 0.3 1.1

(0.1⫾0.0) (0.4⫾0.2) (0.1⫾0.0) (0.3⫾0.1) (0.1⫾0.0) (0.5⫾0.1) (0.1⫾0.0) (0.4⫾0.2)

Cranio-caudal 1.4 4.8 0.7 2.3 0.5 4.2 0.6 2.9

(0.3⫾0.1) (1.1⫾0.3) (0.2⫾0.1) (1.0⫾0.1) (0.1⫾0.1) (1.6⫾0.4) (0.4⫾0.2) (1.6⫾0.5)

Anterior-posterior 2.6 5.1 0.6 2.5 0.6 8.0 0.4 3.3

(0.2⫾0.2) (1.5⫾0.5) (0.1⫾0.0) (1.0⫾0.1) (0.1⫾0.0) (2.3⫾0.6) (0.2⫾0.3) (1.8⫾0.2)