Download daily electronic portal imaging for morbidly

Int. J. Radiation Oncology Biol. Phys., Vol. 59, No. 1, pp. 6 –10, 2004
Copyright © 2004 Elsevier Inc.
Printed in the USA. All rights reserved
0360-3016/04/$–see front matter
doi:10.1016/j.ijrob.2003.12.027
RAPID COMMUNICATION
DAILY ELECTRONIC PORTAL IMAGING FOR MORBIDLY OBESE MEN
UNDERGOING RADIOTHERAPY FOR LOCALIZED PROSTATE CANCER
LAURA ELLEN MILLENDER, M.D.,* MICHELE AUBIN, M.SC.,* JEAN POULIOT, PH.D.,*
KATSUTO SHINOHARA, M.D.,† AND MACK ROACH III, M.D.*†
Departments of *Radiation Oncology and †Urology, University of California San Francisco, Comprehensive Cancer Center,
San Francisco, CA
Purpose: We summarize our experience with a series of morbidly obese men treated using daily online portal
imaging and implanted gold markers to guide external beam radiation therapy (EBRT).
Methods and Materials: Three consecutive morbidly obese men were treated with EBRT for localized prostate
cancer. Daily electronic portal imaging was used to verify patient position. The magnitude and direction of
patient positioning error were documented for each fraction.
Results: The absolute magnitude of positioning error was greatest in the left-right direction with a mean of 11.4
mm/fraction (median, 8 mm; range, 0 – 42 mm). Mean error in the superior-inferior direction was also substantial
at 7.2 mm/fraction (median, 5 mm; range, 0 – 47 mm). Anteroposterior error was the least problematic with a
mean value of 2.6 mm/fraction (median, 2.5 mm; range, 0 – 8 mm).
Conclusions: Daily electronic portal imaging combined with gold fiducial markers dramatically improves the
precision of EBRT in the treatment of morbidly obese men with prostate cancer. Setup error rather than organ
motion appears to be the dominant force in positioning error in obese men. © 2004 Elsevier Inc.
Obesity, Patient positioning, EPID, Radiopaque markers.
standard weekly portal imaging fails to correct for daily
patient positioning error. Ultrasound imaging is limited by
user variability (3), and poor image quality is associated
with increased distance from the abdominal surface to the
isocenter and increased thickness of tissue anterior to the
bladder (4).
Patient setup verification and correction based on bony
anatomy with the use of an online portal imaging device has
substantially reduced setup errors in the case of a 150-kg
patient (5). Daily use of an electronic portal imaging device
(EPID) in combination with implanted gold seed fiducial
markers has been shown to be an accurate way to verify
prostate position during EBRT (6 –10) of the general prostate patient population. This option allows for correction of
organ motion as well as daily setup error. We evaluated the
daily patient positioning error in three consecutive morbidly
obese patients using daily electronic portal imaging and
gold seed fiducial markers to guide therapy.
INTRODUCTION
Morbid obesity, defined as body mass index greater than
40, is associated with severe health complications and a
twofold higher risk for all-cause mortality (1). The prevalence of morbid obesity in the United States has more
than doubled over the last 10 years, with rates of morbid
obesity in American men recently reported at 1.5–3% (1,
2). Given the incidence of prostate cancer in American
men and the prevalence of morbid obesity, it can be
expected that some men with prostate cancer will also be
morbidly obese.
Treatment of this set of patients poses some unique
challenges. These men are frequently not good candidates
for general anesthesia, and surgical procedures such as
radical prostatectomy and permanent or temporary brachytherapy are technically difficult. External beam radiation
therapy (EBRT) is also challenging in obese patients. There
are practical limitations of radiation therapy equipment such
as table weight limits and computed tomography (CT) scan
aperture limits. Daily setup is difficult because skin tattoos
have increased mobility relative to bony structures, making
them less reliable. Commonly used immobilization devices
do not address lateral shifting of the lower abdomen, and
METHODS AND MATERIALS
Three consecutive morbidly obese patients received
EBRT for treatment of localized prostate cancer at the
Reprint requests to: Mack Roach III, M.D., Department of
Radiation Oncology, University of California, San Francisco,
Comprehensive Cancer Center, 1600 Divisadero Street, Box 1708,
San Francisco, CA 94143-1708. Tel: (415) 353-7175; Fax: (415)
353-9883; E-mail: [email protected]
Received Nov 17, 2003, and in revised form Dec 18, 2003.
Accepted for publication Dec 19, 2003.
6
Daily EPID for obese prostate cancer patients
University of California San Francisco between April 2001
and March 2003. All were entered as part of a prospective
Internal Review Board–approved protocol to monitor organ
motion and setup error using daily electronic portal imaging
and gold marker seeds. All patients had T1 or T2 disease
and Gleason scores less than or equal to 7. Maximum
pretreatment prostate specific antigen levels ranged from
8.1 to 45.8 ng/mL.
All patients received EBRT in 180 cGy fractions to a
prescription dose of 7200 cGy (maximum dose range,
7774 – 8000 cGy). All patients were treated with wholepelvis fields followed by a cone down to the prostate and
seminal vesicles and a boost to the prostate alone. Androgen-suppression therapy was initiated in all patients
before radiation therapy and continued during radiation.
Each patient underwent transrectal ultrasound-guided
gold seed placement in the University of California San
Francisco Department of Urology. Gold seeds were
placed during radiation therapy before the first cone
down in 2 patients and before the start of radiation
therapy in 1 patient. One seed was placed in the prostatic
apex and two seeds in the base for a total of three seeds
per patient. Retrograde urethrogram and noncontrast CT
scan were performed on each patient at the time of
simulation. Three-dimensional treatment planning with
Pinnacle software was used in all patients. Each gold
marker (dimensions, 3 mm ⫻ 1.1 mm, Alpha Omega
Services Inc., Bellflower, CA) was identified on the planning CT scans, contoured in the planning software, and
clearly visible on digitally reconstructed radiographs,
simulation films, and portal images.
Left lateral and anteroposterior (AP) portal images were
acquired with 2 cGy each day before treatment using an
amorphous silicon flat panel portal imager (11). The magnitude of patient positioning error was determined in three
directions: superior-inferior (SI), left-right (LR), and AP.
Bony landmarks were used to determine setup error on the
whole pelvis field and gold seed markers were used to
determine patient positioning error on the cone down and
boost fields. Couch moves were made as necessary to align
the pelvis or prostate appropriately within the treatment
field. All moves were made by the radiation technologists
under the guidance of the attending physician. The magnitude and direction of organ motion error was recorded as a
separate endpoint in the patient with gold seeds placed
before the start of radiation therapy.
RESULTS
On 80% of treatment days, a patient move was necessary
to correct for positioning error. Patient moves were considered mandatory if the magnitude of error was greater than 4
mm in any one direction. For the majority of fractions, a
patient move was required in one direction, but on 39% of
treatment days repositioning was needed in two or more
directions. The magnitude of setup error exceeded 20 mm
on 20% of treatment days.
● L. E. MILLENDER et al.
7
Table 1. Absolute daily patient positioning error
Mean
Median
Range
95% CI
SI (mm)
LR (mm)
AP (mm)
7.2
5
0–47
5.3–9.1
11.4
8
0–42
9.0–13.8
2.6
2.5
0–8
1.8–3.3
Abbreviations: SI ⫽ superior/inferior; LR ⫽ left/right; AP ⫽
anteroposterior; CI ⫽ confidence interval.
Table 1 summarizes the magnitude of patient positioning error in each direction. The most pronounced direction of positioning error was LR. Mean daily lateral error
was 11.4 mm for all patients considered as a group. The
median error was 8 mm, and the range was 0 – 42 mm.
Error in the lateral direction exceeded 10 mm on 44% of
treatment days, as shown in Fig. 1. SI positioning error
was also large, with a mean of 7.2 mm per day, median
of 5 mm, and range 0 – 47 mm. Most commonly SI error
was less than 5 mm per day, but on 20% of treatment
visits the magnitude of error in the SI direction exceeded
10 mm. AP error was the least problematic, with a mean
of 2.6 mm and median of 2.5 mm. The magnitude of AP
error did not exceed 10 mm on any treatment visit, and
exceeded 4 mm only 23% of the time.
The nature and magnitude of positioning error for 2
patients are shown in Figs. 2 and 3. Figure 2 shows a
sequence of portal images acquired immediately before
treatment. These images were acquired using 18 MV photons and a 2 monitor unit exposure. The patient was positioned on the treatment couch using skin marks aligned with
the treatment room lasers. The first image showed large
setup error with no bony landmarks clearly identifiable
within the radiation field. After correction, a second image
showed SI setup error requiring the patient to be moved in
the superior direction. During this translation, a new rotation error was introduced. This process was continued until
adequate alignment was achieved, and only then was the
patient treated. The global displacement of the isocenter
between image 5 (correct position) and image 1 (original
position based on skin marks) approached 5 cm. For subsequent fractions, the technologists learned to align the
patient using two or three portal images per day. Before the
prostate boost, three radiopaque markers were implanted to
facilitate alignment and take organ motion into account.
These markers were visible at all incidence angles using a 2
monitor unit exposure. Figure 3 illustrates the variability of
positioning error in a different study patient over the course
of treatment. Magnitude and direction of SI, LR, and AP
error are documented for each of 40 fractions for a single
patient.
Prostate organ motion was recorded as a separate endpoint for 1 patient. Gold seeds were placed before the
start of radiation therapy in this patient. Bony anatomy
and gold seeds were both visible on whole pelvis images,
with three exceptions. Gold seeds were not visible on one
8
I. J. Radiation Oncology
● Biology ● Physics
Volume 59, Number 1, 2004
Fig. 1. Magnitude of patient positioning error by fraction in millimeters (mm) for all patients.
AP portal image because of the magnitude of setup error
and on two lateral portal images because of poor image
quality. Gold seeds were identifiable on all cone down
and boost portal images, but bony landmarks were not
clearly visible on the majority of these images. Bony
landmarks were insufficient for patient alignment on 10
Fig. 2. Sequence of portal images acquired prior treatment delivery with corresponding simulation image for comparison.
Daily EPID for obese prostate cancer patients
● L. E. MILLENDER et al.
9
Fig. 3. Magnitude and direction of daily positioning error in 1 patient as an example of the variation in daily setup of
an obese patient. Bony landmarks were used to determine setup error for fractions 1–25. Gold seeds were used to
determine patient positioning error for fractions 26 – 40 combining organ motion and setup error into one numerical
value.
AP and 14 left lateral cone down and boost images. Mean
magnitudes of organ motion in the SI, LR, and AP
directions were 1.4, 1.8, and 1.5 mm/fraction, and median
magnitudes of organ motion were 1, 2, and 1 mm/fraction, respectively. Because bony anatomy and gold seeds
were not visible on each image, it was impossible to
determine organ motion for each fraction.
No patient experienced any Grade 3 toxicity from seed
placement or radiation therapy. The reported side effects of
seed placement were mild to moderate discomfort at the
time of the procedure and minor rectal bleeding less severe
than that commonly seen with prostate biopsy. The reported
side effects of radiation therapy were mild and included
loose stools and increased urinary obstructive symptoms.
Prior studies have suggested that significant seed migration
should not be expected during radiation therapy (9), and no
obvious seed migration was seen in this patient group,
although this was not specifically evaluated as a separate
endpoint. Total daily treatment time using this technique is
estimated to be 20 min.
DISCUSSION
Setup error in men receiving radiation therapy for treatment
of prostate cancer has been previously documented to average
1– 4 mm/fraction (6, 7, 10, 12). The results of this study clearly
demonstrate that setup error in the SI and LR directions in
morbidly obese men substantially exceeds these average values. Maximum expected daily prostate organ motion ranges
from 1.7 mm/fraction in the LR direction to 4.5 mm/fraction in
the AP direction (7), and in the majority of prostate cancer
patients, prostate organ motion error exceeds setup error (6). In
contrast to the average prostate cancer patient, setup error
rather than organ motion appears to be the dominant force in
positioning error in obese men.
The use of daily electronic portal imaging is beneficial
for proper positioning of morbidly obese prostate cancer
patients. Error correction based on bony landmarks is
adequate for whole pelvis fields where multiple bony
structures can be visualized. When small volumes are
treated with conformal plans, implanted gold seed markers become important for proper patient positioning.
Gold seed markers are more clearly visible than bony
landmarks in small portal images, and they allow for the
simultaneous correction of setup error and organ motion
error. Implanted gold marker seeds maximize the potential benefit of daily on line portal imaging.
Proper patient positioning is critical to gain maximum
benefit of modern treatment planning techniques such as
three-dimensional conformal and intensity-modulated radiation therapy. Because of practical anatomic limitations it is
doubtful that ultrasound-based approaches will be suitable
solutions in this patient population. Given the low risks of
this method of treatment and the substantial amount of setup
error now documented in this group of patients, the daily
use of EPID in combination with gold marker seeds should
be strongly considered in all morbidly obese men receiving
radiation therapy for treatment of prostate cancer. This
technique clearly has the potential to be more broadly
applied, but additional studies should be done before specific recommendations can be made. These findings are so
compelling that we felt it was important to make the radiotherapy community aware of this potential problem despite
the small sample size of this study.
10
I. J. Radiation Oncology
● Biology ● Physics
Volume 59, Number 1, 2004
REFERENCES
1. Freedman DS, Khan LK, Serdula MK, et al. Trends and
correlates of class 3 obesity in the United States from 1990
through 2000. JAMA 2002;288:1758–1761.
2. Flegal KM, Carroll MD, Ogden CL, et al. Prevalence and
trends in obesity among US adults, 1999 –2000. JAMA 2002;
288:1723–1727.
3. 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.
4. 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.
5. Luchka K, Shalev S. Pelvic irradiation of the obese patient: a
treatment strategy involving megavoltage simulation and intratreatment setup corrections. Med Phys 1996;23:1897–1902.
6. 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.
7. 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.
8. Litzenberg D, Dawson LA, Sandler H, et al. Daily prostate
targeting using implanted radiopaque markers. Int J Radiat
Oncol Biol Phys 2002;52:699–703.
9. Pouliot J, Aubin M, Langen K, et al. (Non)-migration of
radiopaque markers used for on-line localization of the prostate with an electronic portal imaging device. Int J Radiat
Oncol Biol Phys 2003;56:862–866.
10. Vigneault E, Pouliot J, Laverdiere J, et al. Electronic portal
imaging device detection of radioopaque markers for the evaluation of prostate position during megavoltage irradiation: A
clinical study. Int J Radiat Oncol Biol Phys 1996;37:205–212.
11. Pouliot J, Aubin M, Chuang C, et al. Clinical use of a flat
panel for megavoltage portal imaging at UCSF [Abstract].
Med Phys 2001;28:1218.
12. Verhey LJ. Immobilizing and positioning patients for radiotherapy. Semin Radiat Oncol 1995;5:100–114.