- Email: [email protected]
A dynamic match-line wedge
Int J. Radmtim
ELSEVIEK
0
Oncology
Bid.
Phy\., Vol. 35. No. I. pp. 161 I h3. lYY6 Copyright C IYYh Elevier Scvmcc Inc. Printed m the IJSA. All rights rrser\ed OM-3016/Yh $1 s.00 + .oo
SSDI: 0360-3016(95)02190-6
Technical
Innovations
and Notes
A DYNAMIC HOBART Roger Williams
SHACKFORD, M.S.
Medical
Center,
Department
MATCH-LINE AND
WEDGE
BENGT E. BJARNGARD, PH.D.
of Radiation
Oncology, Brown University.
Providence.
RI
Purpose: To implement match-line wedges at the abutting edges of x-ray fields using dynamic collimation. Methods and Materials: Fxperiments were made using a computer-controlled linear accelerator equipped with developmental software that allows for collimator jaw motion while the beam is on. The jaws defining the abutting field edges were programmed to move from 1.5 cm inside to 1.5 cm outside the prescribed field during irradiation. Films were taken in plastic phantoms to assess the resulting edge gradient and to evaluate the sensitivity of this technique to setup errors. Results: The measured edge gradient for a single field was 30% per cm. Parallel-opposed lateral fields produced a gradient of 28% per cm along their midline. A simulated central nervous system irradiation with cranial and spinal fields kept dose variations in the field-match region to less than 10% with setup errors of 3 mm. Conclusion: The use of collimator motion during irradiation is an effective and simple means of reducing the dose variation in a field-match region due to setup errors and system tolerances. Treatment time is not increased and labor savings can be achieved when compared to feathering techniques commonly used. Dynamic,
Wedge,
Match-line,
Field matching,
Abutting
fields.
The experiments were made using an accelerator ’ with developmental software that allows for collimator motion while the beam is on. In this case. the jaw defining the match region edge was moved a short distance while the accelerator controlled the jaw speed and dose rate to deliver the beam monitor units with a linear gradient over the region of jaw motion. The amount of jaw travel was
selected using the criteria of Fraass et (11.(2 ). If a setup error of 0.3 cm in field matching is to cause a dose error less than 10%. the dose gradient must be less than 337~ per cm at the matching held edges. This gradient was achieved with a jaw motion of 3 cm. It was advantageous to start with the nominal field size so that the light-field edge could be set at the nominal match-line. The control computer was programmed to move this jaw 1.S cm inside this edge before irradiation began. The jaw was then moved 3 cm outward at a constant rate (monitor units per cm) to end 1.5 cm outside the nominal match-line. The setup for irradiation of the central nervous system (CNS ), shown in Fig. I. with cranial and spinal fields (7) was simulated. The collimator assembly was rotated for the lateral cranial beams to the angle of divergence of the cephalad edge of the posterior spinal held. No table kick or beam splitting was used: thus. the three beam edges in the match region were not coplanar but intersected along a line in the sagittal midplane. A phantom was assembled from rectangular acrylic blocks of various thicknesses to resemble the volume of the head and the neck. Dosimetry films’ were placed
Presented as a poster at the AAPM Annual Meeting, Calgary, Canada, August 1992. Reprint requests to:Hobart Shackford, MS., Roger Williams of Radiation Oncology. 825 Medical Center, Department
Chalkstone Ave., Providence, RI 02YOX. Accepted for publication 30 October 1995. ’ SL2.5. Phillips Medical Systems, Crawley, England. ’ Type V. Eastman Kodak Company, Rochester, NY.
INTRODUCTION
A match-line wedge can be of great value when one abuts two x-ray fields. By inserting a small wedge at the matching edges of the fields, the probability of an overdose or underdose due to positioning errors is much reduced (2, 6). Feathering accomplishes the same by moving the edges a few times during the treatment course. The equivalent of a match-line wedge can be produced with computer-controlled collimator motion during irradiation, a variation on the dynamic wedge ( I, 3-S). The purpose of this study was to evaluate this concept.
METHODS
AND
MATERIALS
162
I. J. Radiation Oncology l Biology l Physics
Volume 35, Number
1, 1996
2
Transverse Distance (cm) a -2
-6 -8 ”
15
10
3
Axial
20
Distance (cm)
Fig. 3. Isodose plot in a coronal plane at 5 cm depth with 3 cm match-line wedges for each field. The isodose levels are between 90 and 115% in 5% steps. The dashed lines represent the phantom sides and the cranial region is to the right. The
central axes of the lateral fields are shown by the vertical dashdot line.
Fig. 1. Posterior (top) and lateral views of the setup for irradiation of the central nervous system that was simulated for this study. The three beam edges in the match region were not coplanar but intersected along a line in the sagittal midplane. The collimator assembly was rotated for the lateral cranial beams about a central axis located behind the eyes to the angle of divergence of the cephalad edge of the posterior spinal field.
between the blocks. A calibration film was exposed with the other films in each experiment and used to convert optical density to dose. Profiles were acquired by manually scanning the films using a digital densitometer, and isodose charts were produced using a film digitizer. Three phantom setups were used. The first setup was one in which the three fields were designed to abut along a line in the sagittal midplane. Using the lateral light field, a match line was drawn on the phantom neck and the posterior light field edge was aligned to the match line. In the second setup, the phantom was shifted off the match line 0.3 cm inferiorly between the cranial and spinal field irradiations to give a gap and simulate a 0.3 cm setup error. For the third setup the phantom was shifted 0.3 cm superiorly to give an overlap and simulate a setup error
140
120
Relative Dose (%)
Relative Dose (??)
100
80 60
-1
0
-2 I
2
Distance (cm) Fig. 2. Dynamic match-line wedge penumbra along the midline of a CNS phantom in the axial direction at 5 cm depth for a single spinal field (solid) and combined parallel-opposed lateral cranial fields (dashed). The dotted line is the static field penumbra.
-I
0
1
2
3
Distance from skin match-line (cm)
Fig. 4. Profiles along the midline (0 cm transverse on Fig. 3) at 5 cm depth with (solid) and without (dashed) match-line wedges. The three curves in each set are for setups of a calculated abutting between the spinal and lateral fields (center), a 0.3 cm overlap (upper), and a 0.3 cm gap (lower).
A dynamic
in the opposite direction setups the second lateral cally by the accelerator.
match-line
wedge l
from the second setup, In all field was positioned automati-
RESULTS Figure 2 shows the dose variation of field edges along a line in the midplane 5 cm below the neck surface, measured with horizontal films. With the match-line wedge, the gradient for the spinal field edge is 30% per cm and for the parallel-opposed lateral fields 28% per cm. For comparison, without a match-line wedge, the penumbra gradient is about 165% per cm. Figure 3 shows an isodose plot of a horizontal film at a 5 cm depth in the phantom for the three-field CNS irradiation with match-line wedges. Figure 4 shows profiles in the axial direction along the phantom midline with and without a match-line wedge and for three phantom setups. The match-line wedge dose variation in the match region changed approximately 15% with the 0.6 cm change in field separation, while the conventional static
H.
SHACKFOKII
ANI)
B. E. BJ;~KNGARD
163
field setup went from an overdosing to a significant underdosing condition. From the static held results it appears that the designed field abutment actually resulted in an overlap of about 0.1 cm. The best dose distribution for the match-line wedge was with a gap of about 0.2 cm.
CONCLUSION Conventional field matching is very sensitive to setup errors and machine tolerances, as seen in Fig. 4. The use of collimator motion during irradiation is a simple means of widening the beam penumbra, making the technique tolerant of setup errors. For the accelerator system used, with 3 cm of jaw motion to create a match-line wedge, the dose variations across the match-line region change by less than 10% when errors typical of careful clinical setup occur. For the three-field setup simulated a matchline gap of about 0.2 cm gives the best dose distribution across the match-line region. The treatment time is not increased and the routine can be installed as a fixed technique on the accelerator menu of beams.
REFERENCES Convery, D. J.; Rosenbloom, M. E. The generation of intensity-modulated fields for conformal radiotherapy by dynamic collimation. Phys. Med. Biol. 3:1359- 1374; 1992. Fraass, B. A.; Tepper, J. E.; Glatstein, E.; Van De Geijn, J. Clinical use of a match-line wedge for adjacent megavoltage radiation field matching. Int. J. Radiat. Oncol. Biol. Phys. 9:209-216; 1983. Kijewski, P. K.; Chin, L. M.; Bjmgard, B. E. Wedge-shaped dose distributions by computer-controlled collimator motion. Med. Phys. 5:426-429; 1978. Lane, R. G.; Loyd, M. D.; Chow, C. H.; Ekwelundu, E.; Rosen, I. I. Improved dose homogeneity in the head and
neck using computer controlled radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 19:1531- 1538; 1990. Leavitt, D. D.; Martin, M.; Moeller, J. H.; Lee, W. L. Dynamic wedge tield techniques through computer-controlled collimator motion and dose delivery. Med. Phys. 17:87-91; 1990. Sohn, J. W.; Schell, M. C.; Dass, K. K.; Sub, J. H.; Tefft, M. Uniform irradiation of the craniospinal axis with a penumbra modifier and an asymmetric collimator. Int. J. Radiat. Oncol. Biol. Phys. 29: 187- 190; 1994. Tatcher, M.; Glicksman, A. S. Field matching considerations in craniospinal irradiation. Int. J. Radiat. Oncol. Biol. Phys. 17:865-869; 1989.
ELSEVIEK
0
Oncology
Bid.
Phy\., Vol. 35. No. I. pp. 161 I h3. lYY6 Copyright C IYYh Elevier Scvmcc Inc. Printed m the IJSA. All rights rrser\ed OM-3016/Yh $1 s.00 + .oo
SSDI: 0360-3016(95)02190-6
Technical
Innovations
and Notes
A DYNAMIC HOBART Roger Williams
SHACKFORD, M.S.
Medical
Center,
Department
MATCH-LINE AND
WEDGE
BENGT E. BJARNGARD, PH.D.
of Radiation
Oncology, Brown University.
Providence.
RI
Purpose: To implement match-line wedges at the abutting edges of x-ray fields using dynamic collimation. Methods and Materials: Fxperiments were made using a computer-controlled linear accelerator equipped with developmental software that allows for collimator jaw motion while the beam is on. The jaws defining the abutting field edges were programmed to move from 1.5 cm inside to 1.5 cm outside the prescribed field during irradiation. Films were taken in plastic phantoms to assess the resulting edge gradient and to evaluate the sensitivity of this technique to setup errors. Results: The measured edge gradient for a single field was 30% per cm. Parallel-opposed lateral fields produced a gradient of 28% per cm along their midline. A simulated central nervous system irradiation with cranial and spinal fields kept dose variations in the field-match region to less than 10% with setup errors of 3 mm. Conclusion: The use of collimator motion during irradiation is an effective and simple means of reducing the dose variation in a field-match region due to setup errors and system tolerances. Treatment time is not increased and labor savings can be achieved when compared to feathering techniques commonly used. Dynamic,
Wedge,
Match-line,
Field matching,
Abutting
fields.
The experiments were made using an accelerator ’ with developmental software that allows for collimator motion while the beam is on. In this case. the jaw defining the match region edge was moved a short distance while the accelerator controlled the jaw speed and dose rate to deliver the beam monitor units with a linear gradient over the region of jaw motion. The amount of jaw travel was
selected using the criteria of Fraass et (11.(2 ). If a setup error of 0.3 cm in field matching is to cause a dose error less than 10%. the dose gradient must be less than 337~ per cm at the matching held edges. This gradient was achieved with a jaw motion of 3 cm. It was advantageous to start with the nominal field size so that the light-field edge could be set at the nominal match-line. The control computer was programmed to move this jaw 1.S cm inside this edge before irradiation began. The jaw was then moved 3 cm outward at a constant rate (monitor units per cm) to end 1.5 cm outside the nominal match-line. The setup for irradiation of the central nervous system (CNS ), shown in Fig. I. with cranial and spinal fields (7) was simulated. The collimator assembly was rotated for the lateral cranial beams to the angle of divergence of the cephalad edge of the posterior spinal held. No table kick or beam splitting was used: thus. the three beam edges in the match region were not coplanar but intersected along a line in the sagittal midplane. A phantom was assembled from rectangular acrylic blocks of various thicknesses to resemble the volume of the head and the neck. Dosimetry films’ were placed
Presented as a poster at the AAPM Annual Meeting, Calgary, Canada, August 1992. Reprint requests to:Hobart Shackford, MS., Roger Williams of Radiation Oncology. 825 Medical Center, Department
Chalkstone Ave., Providence, RI 02YOX. Accepted for publication 30 October 1995. ’ SL2.5. Phillips Medical Systems, Crawley, England. ’ Type V. Eastman Kodak Company, Rochester, NY.
INTRODUCTION
A match-line wedge can be of great value when one abuts two x-ray fields. By inserting a small wedge at the matching edges of the fields, the probability of an overdose or underdose due to positioning errors is much reduced (2, 6). Feathering accomplishes the same by moving the edges a few times during the treatment course. The equivalent of a match-line wedge can be produced with computer-controlled collimator motion during irradiation, a variation on the dynamic wedge ( I, 3-S). The purpose of this study was to evaluate this concept.
METHODS
AND
MATERIALS
162
I. J. Radiation Oncology l Biology l Physics
Volume 35, Number
1, 1996
2
Transverse Distance (cm) a -2
-6 -8 ”
15
10
3
Axial
20
Distance (cm)
Fig. 3. Isodose plot in a coronal plane at 5 cm depth with 3 cm match-line wedges for each field. The isodose levels are between 90 and 115% in 5% steps. The dashed lines represent the phantom sides and the cranial region is to the right. The
central axes of the lateral fields are shown by the vertical dashdot line.
Fig. 1. Posterior (top) and lateral views of the setup for irradiation of the central nervous system that was simulated for this study. The three beam edges in the match region were not coplanar but intersected along a line in the sagittal midplane. The collimator assembly was rotated for the lateral cranial beams about a central axis located behind the eyes to the angle of divergence of the cephalad edge of the posterior spinal field.
between the blocks. A calibration film was exposed with the other films in each experiment and used to convert optical density to dose. Profiles were acquired by manually scanning the films using a digital densitometer, and isodose charts were produced using a film digitizer. Three phantom setups were used. The first setup was one in which the three fields were designed to abut along a line in the sagittal midplane. Using the lateral light field, a match line was drawn on the phantom neck and the posterior light field edge was aligned to the match line. In the second setup, the phantom was shifted off the match line 0.3 cm inferiorly between the cranial and spinal field irradiations to give a gap and simulate a 0.3 cm setup error. For the third setup the phantom was shifted 0.3 cm superiorly to give an overlap and simulate a setup error
140
120
Relative Dose (%)
Relative Dose (??)
100
80 60
-1
0
-2 I
2
Distance (cm) Fig. 2. Dynamic match-line wedge penumbra along the midline of a CNS phantom in the axial direction at 5 cm depth for a single spinal field (solid) and combined parallel-opposed lateral cranial fields (dashed). The dotted line is the static field penumbra.
-I
0
1
2
3
Distance from skin match-line (cm)
Fig. 4. Profiles along the midline (0 cm transverse on Fig. 3) at 5 cm depth with (solid) and without (dashed) match-line wedges. The three curves in each set are for setups of a calculated abutting between the spinal and lateral fields (center), a 0.3 cm overlap (upper), and a 0.3 cm gap (lower).
A dynamic
in the opposite direction setups the second lateral cally by the accelerator.
match-line
wedge l
from the second setup, In all field was positioned automati-
RESULTS Figure 2 shows the dose variation of field edges along a line in the midplane 5 cm below the neck surface, measured with horizontal films. With the match-line wedge, the gradient for the spinal field edge is 30% per cm and for the parallel-opposed lateral fields 28% per cm. For comparison, without a match-line wedge, the penumbra gradient is about 165% per cm. Figure 3 shows an isodose plot of a horizontal film at a 5 cm depth in the phantom for the three-field CNS irradiation with match-line wedges. Figure 4 shows profiles in the axial direction along the phantom midline with and without a match-line wedge and for three phantom setups. The match-line wedge dose variation in the match region changed approximately 15% with the 0.6 cm change in field separation, while the conventional static
H.
SHACKFOKII
ANI)
B. E. BJ;~KNGARD
163
field setup went from an overdosing to a significant underdosing condition. From the static held results it appears that the designed field abutment actually resulted in an overlap of about 0.1 cm. The best dose distribution for the match-line wedge was with a gap of about 0.2 cm.
CONCLUSION Conventional field matching is very sensitive to setup errors and machine tolerances, as seen in Fig. 4. The use of collimator motion during irradiation is a simple means of widening the beam penumbra, making the technique tolerant of setup errors. For the accelerator system used, with 3 cm of jaw motion to create a match-line wedge, the dose variations across the match-line region change by less than 10% when errors typical of careful clinical setup occur. For the three-field setup simulated a matchline gap of about 0.2 cm gives the best dose distribution across the match-line region. The treatment time is not increased and the routine can be installed as a fixed technique on the accelerator menu of beams.
REFERENCES Convery, D. J.; Rosenbloom, M. E. The generation of intensity-modulated fields for conformal radiotherapy by dynamic collimation. Phys. Med. Biol. 3:1359- 1374; 1992. Fraass, B. A.; Tepper, J. E.; Glatstein, E.; Van De Geijn, J. Clinical use of a match-line wedge for adjacent megavoltage radiation field matching. Int. J. Radiat. Oncol. Biol. Phys. 9:209-216; 1983. Kijewski, P. K.; Chin, L. M.; Bjmgard, B. E. Wedge-shaped dose distributions by computer-controlled collimator motion. Med. Phys. 5:426-429; 1978. Lane, R. G.; Loyd, M. D.; Chow, C. H.; Ekwelundu, E.; Rosen, I. I. Improved dose homogeneity in the head and
neck using computer controlled radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 19:1531- 1538; 1990. Leavitt, D. D.; Martin, M.; Moeller, J. H.; Lee, W. L. Dynamic wedge tield techniques through computer-controlled collimator motion and dose delivery. Med. Phys. 17:87-91; 1990. Sohn, J. W.; Schell, M. C.; Dass, K. K.; Sub, J. H.; Tefft, M. Uniform irradiation of the craniospinal axis with a penumbra modifier and an asymmetric collimator. Int. J. Radiat. Oncol. Biol. Phys. 29: 187- 190; 1994. Tatcher, M.; Glicksman, A. S. Field matching considerations in craniospinal irradiation. Int. J. Radiat. Oncol. Biol. Phys. 17:865-869; 1989.