- Email: [email protected]
Comprehensive Monte Carlo calculations of AAPM Task Group Report No. 43 dosimetry parameters for the Model 3500 I-Plant 125I brachytherapy source
Applied Radiation and Isotopes 57 (2002) 381–389
Comprehensive Monte Carlo calculations of AAPM Task Group Report No. 43 dosimetry parameters for the Model 3500 I-Plant 125I brachytherapy source Mark J. Rivard* Department of Radiation Oncology, Tufts University School of Medicine, New England Medical Center #246, 750 Washington Street, Boston, MA 02111, USA Received 29 October 2001; received in revised form 2 January 2002; accepted 7 March 2002
Abstract The Model 3500 I-Plant 125I brachytherapy source, produced by Implant Sciences Corporation, is created through a novel technique which utilizes ion implantation of 124Xe into a ceramic matrix followed by neutron activation of the substrate. Clinical dosimetry parameters were calculated using MCNP. For data processing, a 0.376 cm active length was used for the geometry function. The 2-D anisotropy function, F ðr; yÞ; exhibited less anisotropy than the Amersham Health Model 6711 source, and decreased as radial distance, r, increased. The anisotropy constant, f% an ; was 0.933; dose rate constant, L; was 1.01770.005 cGy/h/U; and radial dose function, gðrÞ; for r={0.1, 0.2, 0.5, 2, 5, 10 cm} was 0.990, 1.014, 1.030, 0.872, 0.448, and 0.114, respectively. In general, there was good agreement with experimental results recently obtained by others using thermoluminescent dosimeters. r 2002 Elsevier Science Ltd. All rights reserved. Keywords: Monte Carlo calculations;
125
I; Brachytherapy; Dosimetry; TG-43
1. Introduction In August 2000, Implant Science Corporation (ISC) introduced the Model 3500 I-Plant 125I brachytherapy source to market. Both the design and fabrication process for this source are unique in comparison to other 125I brachytherapy sources, with the merits of this approach described by Duggan and Johnson (2001). The American Association of Physicists in Medicine (AAPM) recommends at least one complete set of experimental measurements and at least one complete set of Monte Carlo calculations preceding clinical implementation (Williamson et al., 1998). Duggan and Johnson satisfied the first recommendation obtaining measurements using thermoluminescent dosimeters (TLDs). Wallace (2002) also performed experimental measurements using TLDs, but results from Monte
*Tel.: +1-617-636-1680; fax: +1-617-636-7621. E-mail address: [email protected] (M.J. Rivard).
Carlo calculations were still needed. Therefore, the purpose of this paper is to satisfy the latter recommendation and present clinical brachytherapy dosimetry parameters as described in the AAPM Task Group Report No. 43 (TG-43) by Nath et al. (1995).
2. Materials and methods 2.1. Source geometry, composition, and energy spectrum
125
I photon
The materials and methods employed in this study were similar to those employed in previous studies by Rivard (2001a, b); the reader is referred to these papers for greater detail. A graphical depiction of the Model 3500 Monte Carlo geometry space is presented in Fig. 1. Based on physical examination of the source at various stages of fabrication, the source construction was assumed to be symmetric about the transverse-plane with the internal components (silver marker and quartz
0969-8043/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 9 6 9 - 8 0 4 3 ( 0 2 ) 0 0 1 1 0 - 0
382
M.J. Rivard / Applied Radiation and Isotopes 57 (2002) 381–389
Fig. 1. (a) Model 3500 125I source geometry as modeled using MCNP. Details of the capsule end cap weld, conical silver radiographic marker, and quartz tube are evident. The total capsule length is 4.60 mm. (b) Close-up of Model 3500 125I source geometry detailing the 124 Xe implanted quartz tube. A quartz tube is ion-implanted with 124Xe which is neutron activated to convert 124Xe into 125Xe which decays to 125I. For radiological safety purposes, the active layer is coated with a 5 mm layer of SiO2.
tube) sharing the same origin coordinates and source long-axis as the capsule. The titanium capsule inner diameter (ID) and outer diameter (OD) was 0.7239 and 0.8356 mm, respectively, with an overall length of 4.60 mm and 0.3302 mm thick hemispherical capsule end cap welds. The mass densities of the conical-ended silver marker were 10.50, 4.54 g/cm3 for surgical grade (ASTM F67-95) titanium, and 0.00120 g/cm3 for dry air modeled within the Ti capsule and used for air kerma calculations. Particular to this source is the makeup of the quartz tube (2.21 g/cm3) surrounding the 0.406 mm diameter silver marker. The quartz tube ID and OD were 0.4318 and 0.6350 mm, respectively, with a length of 3.76 mm. Atop this tube was deposited a 16 mm layer of Si into which approximately 1.5 1017 124Xe nuclei were implanted. Over this layer was deposited a 5 mm SiO2 overcoat layer to contain the radioactive 125I product. The outer diameter of the entire tube was 0.6770 mm. Based on this 124Xe loading, a mass density of 2.58 g/cm3 was calculated for the active Si/Xe layer. Following neutron irradiation and decay preceding clinical implementation, the atomic density of the active
Table 1 125 I photon source model Photon energy (keV) 27.202 27.472 30.98 31.71 35.492 28.37 keV avg.
Photons per disintegration (%) 40.6 75.7 20.2 4.39 6.68 147.6% total
Source photons with energies less than 5 keV, and source electrons, were ignored due to their relatively low yields and negligible chance of penetrating the titanium capsule.
layer was predicted to be 97.6174% Si, 2.3804% Xe, 0.0016% I, and 0.0006% Te. Using these atomic and mass densities and source geometry, the appropriate photon (MCPLIB02) and electron (EL1) cross-section libraries were chosen from the DLC-189 library supplied with the MCNP software. The 125I photon spectrum was
M.J. Rivard / Applied Radiation and Isotopes 57 (2002) 381–389
383
Table 2 Geometry function multiplied by the radial distance squared, Gðr; yÞ r2 ; for the Model 3500 I-Plant 125I brachytherapy source using the simple line equation with L ¼ 0:376 cm with Gðr0 ; y0 Þ ¼ 0:9885 r (cm)
01
101
201
301
401
501
601
701
801
901
0.25 0.50 1.00 2.00 5.00 10.00
2.3015 1.1647 1.0366 1.0089 1.0014 1.0004
2.0723 1.1558 1.0351 1.0086 1.0014 1.0003
1.6901 1.1316 1.0306 1.0075 1.0012 1.0003
1.3921 1.0978 1.0238 1.0059 1.0009 1.0002
1.1879 1.0608 1.0157 1.0040 1.0006 1.0002
1.0504 1.0258 1.0074 1.0019 1.0003 1.0001
0.9588 0.9964 0.9998 1.0000 1.0000 1.0000
0.9004 0.9745 0.9937 0.9984 0.9997 0.9999
0.8679 0.9610 0.9898 0.9974 0.9996 0.9999
0.8574 0.9565 0.9885 0.9971 0.9995 0.9999
Table 3 g(r) for the ISC I-Plant Model 3500
125
I brachytherapy source
Model 3500 r (cm) 0.050 0.075 0.100 0.150 0.174 0.200 0.250 0.275 0.300 0.350 0.400 0.450 0.500 0.600 0.700 0.750 0.800 0.900 1.000 1.250 1.500 1.750 2.000 2.250 2.500 2.750 3.000 3.500 4.000 4.500 5.000 5.500 6.000 7.000 7.500 8.000 9.000 10.000
TLDs by Duggan and Johnson (2001)
Model 6711 TLDs by Wallace (2002)
MCNP in this work
Nath et al. (1995)
1.158 1.004 0.990 1.006 1.000
1.026
1.022
1.018
1.000
1.000
0.936
0.935
0.855
0.846
0.770
0.749
0.687 0.609 0.537 0.473 0.416
0.655 0.570 0.498 0.440 0.394 0.357 0.327 0.271
0.320 0.246
0.209 0.149 0.134
1.014 1.021 1.024 1.027 1.028 1.029 1.030 1.030 1.030 1.024 1.021 1.017 1.010 1.000 0.976 0.943 0.908 0.872 0.832 0.795 0.756 0.717 0.643 0.573 0.508 0.448 0.395 0.347 0.265 0.231 0.201 0.151 0.114
1.04
1.00 0.926 0.832 0.731 0.632 0.541 0.463 0.397 0.344 0.300 0.264 0.204
Also shown are ISC Model 3500 results obtained by Duggan and Johnson (2001); Wallace (2002), and Amersham Health Model 6711 results by Nath et al. (1995).
384
M.J. Rivard / Applied Radiation and Isotopes 57 (2002) 381–389
Fig. 2. (a) Model 3500 125I gðrÞ calculated using MCNP, obtained using TLDs by Duggan and Johnson (2001) and Wallace (2002), and for the Amersham Health Model 6711 125I source as reported by Nath et al. (1995) in TG-43. (b) Close-up of the calculated Model 3500 gðrÞ data. It is evident that the dosimetry data normalized to the simple line source geometry function unexpectedly increases for r o 1 mm.The Monte Carlo-derived geometry function, or particle streaming function, permitted normalization of gðrÞ with expected results.
characterized as five photons with energies and abundances listed in Table 1, and emanated isotropically from the active layer. 2.2. Calculation techniques Monte Carlo methods using MCNP v.4B2 in a parallel-processing environment (Briesmeister, 1997;
’ yÞ in Rivard, 1999, 2000) were used to calculate Dðr; water. Dosimetry data in water were calculated at radial distances from the Ti capsule ranging from 0.050 to 10 cm, and over angles ranging from 01 to 1801 using the * F8 tally in a 30 cm diameter phantom. SK was also determined using the *F8 tally in dry air at standard temperature and pressure conditions (221C and 101.325 kPa) in a 6 m diameter phantom as a function
M.J. Rivard / Applied Radiation and Isotopes 57 (2002) 381–389
385
of distance, and by using the product of free-space air ’ free-space ; and the distance, d, squared. kerma rate, KðdÞ ’ free-space results, two corrections To obtain accurate KðdÞ were required:
3. Results and discussion
’ in-air were 1. Energy dependence of contributions to KðdÞ tracked, and contamination by K-edge titanium X’ in-air rays (4.5 and 4.9 keV) were subtracted from KðdÞ ’ in-air results were results. Specifically, the Eg¼4:7 keV KðdÞ ’ in-air subtracted from the Eg¼4:5 keV KðdÞ and ’ Eg¼4:9 keV KðdÞin-air results, these data were then ’ in-air combined and subtracted from the total KðdÞ contribution from all photons at a given distance. ’ in-air results were further corrected for 2. Corrected KðdÞ attenuation and scatter by air by extrapolating ’ free-space : ’ in-air d 2 to zero distance to obtain KðdÞ KðdÞ
The Model 3500 gðrÞ is presented in Table 3. In comparison of Model 3500 results determined using TLDs by Duggan and Johnson (2001) and Wallace (2002) in Fig. 2a, the calculated values grow larger as radius increases. This slight difference may be due to inaccurate source modeling, or may be due to errors in the media correction factor used to convert TLD results in plastic to water. For further comparison, the Model 3500 gðrÞ data are presented with the Amersham Health Model 6711 125I brachytherapy source gðrÞ data by Nath et al. (1995). The Model 3500 radial dose function appears more penetrating than the Model 6711. This increased penetration may be due to proportionally lower silver fluorescence generated in the Model 3500 source relative to the Model 6711 source. A harder photon spectrum like the 6702 is expected due to the increased spacing between the 125I active region and the silver marker due to the quartz tube. The geometric impact of the slight difference in active lengths (3.76 mm used herein vs. 4 mm used by Duggan and Johnson and also by Wallace) would become significant at the 1% and 3% levels at distances of only 3.2 and 1.4 mm along the transverse-plane, respectively. Fig 2b presents gðrÞ data at exceptionally close distances to the Ti capsule. The gðrÞ data increases beneath 1.0 mm because the line source equation no longer accurately models the photon fluence from the tubular active region. Using a Monte Carlo-derived geometry function intrinsically more representative of the photon fluence (F2 or F4), it is clear that the
These corrections are similar to those applied by Hedtj.arn et al. (2000), except here a straight-line was used for the fitting. The geometry function, Gðr; yÞ; was determined using a simple line source equation with L ¼ 0:376 cm and Gðr0 ; y0 Þ ¼ 0:9885: To ensure clarity, example values of Gðr; yÞ r2 are presented in Table 2. Gðr; yÞ was used in ’ yÞ results and data processing to normalize the Dðr; determine the remaining TG-43 parameters: % an : A total of 2 109 source L; gðrÞ; Fðr; yÞ; fan ðrÞ; and f photon histories were processed for each set of in-water and in-air calculations. Subsequent statistical uncertainties (1s) in gðrÞ and fan ðrÞ were typically less than 0.3%. Statistical uncertainties in F ðr; yÞ varied as a function of y; and were typically o 0.3% on the transverse-plane and B3% for yB01 or 1801. The statistical uncertainty in L was based on combined statistical uncertainties ’ 0 ; y0 Þ (0.07%) and KðdÞ ’ free-space from derivation of Dðr % an value (0.49%), and totaled 0.50% (1s). Finally, the f 2 was obtained by using a 1=r –weighting of fan ðr > 1 cmÞ with a 1s error likely less than 0.3%.
3.1. Radial dose function, g(r)
Table 4 ’ in-air output (MeV/g/s.p.) at various distances, relative contribution by 4.5 and 4.9 keV Ti K-edge X-rays, and Monte Carlo KðdÞ ’ in-air d 2 values corrected KðdÞ d (cm)
’ in-air MCNP output (MeV/g/s.p.) KðdÞ
Ti K-edge contribution (%)
’ in-air d 2 (MeV cm2/g/s.p.) Corrected KðdÞ
20 30 50 75 100 125 150 175 200
6.78670.041 10–7 2.98170.016 10–7 1.07370.006 10–7 4.77170.025 10–8 2.69670.014 10–8 1.72470.007 10–8 1.19170.005 10–8 8.67370.036 10–9 6.67470.020 10–9
1.6 0.82 0.32 0.13 0.064 0.031 0.015 0.009 0.006
2.672 10–4 2.661 10–4 2.673 10–4 2.680 10–4 2.694 10–4 2.693 10–4 2.679 10–4 2.656 10–4 2.670 10–4
’ free-space value, was determined by extrapolating to zero distance to remove the minor contribution of The free-space air kerma rate, KðdÞ in-air attenuation and scatter. For derivation of L, the reference dose rate (2.72070.002 10–4 MeV/g/s.p.) was divided by ’ free-space d 2 (2.67670.013 10–4 MeV cm2/g/s.p.) and found to be 1.01770.005 cGy/h/U. KðdÞ
M.J. Rivard / Applied Radiation and Isotopes 57 (2002) 381–389
386
’ 0 ; y0 Þ=ðKðdÞ ’ in-air d 2 ) values as a function of distance are presented, and are corrected by subtraction of titanium Fig. 3. Calculated Dðr K-edge X-ray contributions to in-air kerma strength. Extrapolation to zero distance permitted derivation of the Model 3500 L value1.01770.005 cGy/h/U which was in good agreement with Model 3500 TLD results (1.01 cGy/h/U) obtained by Duggan and Johnson (2001) and by Wallace (2002).
Table 5 F ðr; yÞ derived using Monte Carlo methods for the Model 3500 r (cm) 01 0.050 0.075 0.100 0.150 0.200 0.250 0.500 1.000 2.000 5.000 10.000
0.494 0.610 0.580 0.652 0.690 0.709
1
5
0.531 0.538 0.548 0.596 0.677 0.723
1
10
0.987 0.574 0.513 0.561 0.626 0.700 0.742
1
15
1.128 0.945 0.655 0.592 0.634 0.685 0.746 0.780
1
20
1.039 0.921 0.785 0.679 0.705 0.743 0.789 0.815
25
1
1.107 0.989 0.934 0.858 0.756 0.765 0.791 0.823 0.843
1
1
125
I brachytherapy source
1
30
35
40
451
501
551
601
651
701
751
801
851
1.046 0.978 0.940 0.899 0.808 0.813 0.830 0.854 0.872
1.107 1.010 0.970 0.952 0.930 0.856 0.848 0.863 0.882 0.893
1.050 0.996 0.968 0.960 0.943 0.892 0.885 0.893 0.905 0.912
1.024 0.988 0.974 0.969 0.959 0.919 0.910 0.915 0.924 0.931
1.006 0.990 0.977 0.975 0.967 0.944 0.933 0.934 0.941 0.947
0.998 0.989 0.982 0.977 0.977 0.961 0.950 0.953 0.956 0.957
1.067 0.994 0.993 0.988 0.984 0.986 0.974 0.967 0.967 0.968 0.972
1.027 0.995 0.996 0.989 0.988 0.991 0.983 0.979 0.977 0.979 0.979
0.996 0.996 0.988 0.987 0.988 0.995 0.990 0.987 0.987 0.986 0.990
0.990 0.999 1.001 0.992 0.994 0.994 0.994 0.993 0.993 0.992 0.994
0.985 0.996 0.999 0.996 0.997 1.000 0.997 0.997 0.997 0.996 0.997
0.988 1.003 1.000 0.994 0.998 0.996 1.002 1.001 1.000 0.999 1.001
By definition, F ðr; y ¼ 900 Þ was unity. There was generally greater anisotropy along the source long-axis, towards smaller y angles, and this anisotropy decreased as radial distance increased. Blank data indicate the position was located within the source where radiation dosimetry data are not clinically relevant.
supra-linear gðrÞ effect beneath 1 mm is not evident. Both data sets converge at r > 3 mm. While a more accurate equation for the geometry function was proposed by Rivard (1999), later discussion by Kouwenhoven et al. (2001) and realization of current treatment planning software limitations suggest continued use of the simple point and line equations. Therefore, the Monte Carlo-derived geometry function, with a
proposed nomenclature of the ‘‘particle streaming function’’, may be used in other applications and is not currently recommended for clinical use with conventional brachytherapy treatment planning software. For convenience of clinical implementation, the line source gðrÞ data have been fit to a fifth order polynomial with parameters given below. This equation fit the line source gðrÞ data within an absolute deviation of 71.0%
M.J. Rivard / Applied Radiation and Isotopes 57 (2002) 381–389
over a radial range of 0.1–10 cm. 2
a0 ¼ 1:005; a1 ¼ 9:422 10 ; 1 a2 ¼ 1:217 10 ; a3 ¼ 2:577 102 ; a4 ¼ 2:325 103 ; a5 ¼ 7:828 105 : 3.2. Dose rate constant In terms of Monte Carlo output, the reference dose ’ 0 ; y0 Þ; was 2.72070.002 10–4 MeV/g/s.p. rate, Dðr
387
where s.p. indicates source particle. Monte Carlo ’ in-air output at various distances, relative contribuKðdÞ tion by 4.5 and 4.9 keV Ti K-edge X-rays, and the ’ in-air d 2 values are given in Table 4. corrected KðdÞ After subtraction of titanium K-edge X-ray contribu’ 0 ; y0 Þ=KðdÞ ’ in-air d 2 ) was fit to a tions, the ratio Dðr straight line (y ¼ 1:6737 1026 w þ 1:0166) to obtain ’ free-space : Extrapolation to d ¼ 0 in Fig. 3 ’ 0 ; y0 Þ=KðdÞ Dðr provided a Model 3500 L value of 1.01770.005 cGy/h/U. Duggan and Johnson obtained 1.01 cGy/h/U, as did
Fig. 4. F ðr; yÞ derived using Monte Carlo methods for the Model 3500 125I brachytherapy source. In (b) and (c), the F ðr ¼ 1 cm; yÞ and F ðr ¼ 5 cm; yÞ data are presented, respectively, and indicate fairly good agreement with results obtained by Duggan and Johnson (2001) and by Wallace (2002) for the same type of source. In comparison, the Model 6711 source exhibits greater anisotropy.
388
M.J. Rivard / Applied Radiation and Isotopes 57 (2002) 381–389
Fig. 4 (continued).
Wallace. While Monte Carlo statistical uncertainties were relatively small, uncertainties due to source modeling, cross-section library data, low-energy photon transport physics, or errors in the plastic-to-water correction factors used with TLD results may all contribute to increase the total uncertainty. Compared with the reported Model 6702 and 6711 L values of 1.04 and 0.98 cGy/h/U presented by Williamson et al. (1999), the Model 3500 L was quite similar. This L-averaged behavior was expected since the Model 3500 produces Ag X-rays due to the silver marker (like the Model 6711), yet the 125I is in contact with a relatively low-Z backing (like the Model 6702). 3.3. Anisotropy Table 5 presents calculated Model 3500 F ðr; yÞ data. Results generally indicate that anisotropy increases as the radial distance decreases and as the angle from the transverse-plane increases. Fig. 4a graphically depicts these F ðr; yÞ data. In Figs. 4b and c, F ðr ¼ 1; yÞ and F ðr ¼ 5; yÞ are compared with result by Duggan and Johnson. General agreement is evident here, and with results obtained by Wallace with differences approaching 10% in some instances. Neither of these investigators observed the dip in F ðr; yÞ near yB100 which is likely due to capsule shielding of the source photons by the end cap weld. This is confirmed using simple geometry. Since the proportion of primary-to-scatter dose decreases as distance increases, it is not surprising that the magnitude of the 101 dip decreases as r increases. Upon comparison of Model 3500 F ðr; yÞ data with Model 6702 and 6711 F ðr; yÞ data by Nath et al.,
the Model 3500 125I brachytherapy source appears to have anisotropy more similar to the Model 6702. Table 6 presents calculated Model 3500 fan ðrÞ and fan values. Results differ by B1% with TLD measurements by Duggan and Johnson and by Wallace. The fan value did not significantly vary depending on whether 0.5 or 0.25 cm increments were used in the fan ðr > 1 cmÞ integration. As expected from the F ðr; yÞ data, Model 6711 fan ðrÞ data differ significantly from the Model 3500 fan ðrÞ data, and the Model 3500 125I source exhibits less anisotropy. Considering the combined experimental and calculative uncertainties, the calculated fan for the Model 3500 source was in good agreement with results obtained by the three previously mentioned investigators. Upon dividing each fan ðrÞ value by fan as proposed in Eq. (1) by Rivard (2001b) for a 103Pd source of significantly different construction, these Model 3500 data fit on average within 1% over the entire radial range.
4. Summary The gðrÞ; F ðr; yÞ; fan ðrÞ; and fan values obtained using Monte Carlo calculations and those obtained by others using TLDs for dosimetry of the Model 3500 125I brachytherapy source were in close agreement. Differences due to use of different active source lengths were negligble beyond 1 mm. In comparison to the Model 6711 125I brachytherapy source, the Model 3500 had more isotropic dose distributions, a slightly higher dose rate constant, and slightly increased penetration in water. There was good agreement between the
M.J. Rivard / Applied Radiation and Isotopes 57 (2002) 381–389 Table 6 1-D anisotropy functions, fan ðrÞ; and anisotropy constants, fan ; for the Model 3500 and Model 6711 125I brachytherapy sources Model 3500 r (cm)
0.25 0.30 0.35 0.40 0.45 0.50 0.60 0.70 0.80 0.90 1.00 1.25 1.50 1.75 2.00 2.50 3.00 3.50 4.00 4.50 5.00 6.00 7.00 10.00 fan
Model 6711
TLDs by TLDs by Duggan and Wallace Johnson (2002) (2001)
1.01
0.956
0.96
0.948
0.94
0.944 0.943 0.943 0.944 0.945
0.92
0.948 0.951 0.952
0.98 0.97
0.95
0.95
0.93 0.96
MCNP in this work
1.164 1.077 1.029 1.005 0.985 0.973 0.954 0.945 0.940 0.936 0.933 0.930 0.931 0.931 0.931 0.932 0.934 0.935 0.937 0.938 0.938 0.939 0.942 0.948 0.933
Nath et al.(1995)
0.944
0.936 0.893 0.887 0.884 0.880 0.901
0.93
2
fan was obtained using a 1/r –weighting for fan ðr > 1Þ data.
calculated and measured L values for the Model 3500 125 I brachytherapy source.
Acknowledgements Funding was provided through an unrestricted educational research grant provided by Implant Sciences Corporation (Wakefield, MA).
389
References Briesmeister, J.F., 1997. MCNP—A general Monte Carlo N-particle transport code system, Version 4B, LA12625-M. Duggan, D.M., Johnson, B.L., 2001. Dosimetry of the I-Plant Model 3500 iodine-125 brachytherapy source. Med. Phys. 28, 661–670. Kouwenhoven, E., van der Laars, R., Schaart, D.R., 2001. Variation in interpretation of the AAPM TG-43 geometry factor leads to unclearness in brachytherapy dosimetry. Med. Phys. 28, 1965–1966. Hedtj.arn, H., Carlsson, G.A., Williamson, J.F., 2000. Monte Carlo-aided dosimetry of the Symmetra model I25.S06 I125, interstitial brachytherapy seed. Med. Phys. 27, 1076–1084. Nath, R., Anderson, L.L., Luxton, G., Weaver, K.A., Williamson, J.F., Meigooni, A.S., 1995. Dosimetry of interstitial brachytherapy sources: Recommendations of the AAPM Radiation Therapy Committee Task Group No. 43. Med. Phys. 22, 209–234. Rivard, M.J., 1999. Refinements to the geometry factor used in the AAPM Task Group Report No. 43 necessary for brachytherapy dosimetry calculations. Med. Phys. 26, 2445–2450. Rivard, M.J., 2000. Neutron dosimetry for a general 252Cf brachytherapy source. Med. Phys. 27, 2803–2815. Rivard, M.J., 2001a. Monte Carlo calculations of AAPM Task Group Report No. 43 dosimetry parameters for the MED3631-A/M 125I source. Med. Phys. 28, 629–637. Rivard, M.J., 2001b. A discretized approach to determining TG-43 brachytherapy dosimetry parameters: Case study using Monte Carlo calculations for the MED3633 103Pd source. Appl. Radiat Isot. 55, 775–782. Wallace, R.E., 2002. Model 3500 125I brachytherapy source dosimetric characterization. Appl. Radiat. Isot. 56, 581–587. Williamson, J.F., Coursey, B.M., DeWerd, L.A., Hanson, W.F., Nath, R., 1998. Dosimetric prerequisites for routine clinical use of new low energy photon interstitial brachytherapy sources. Med. Phys. 25, 2269–2270. Williamson, J.F., Coursey, B.M., DeWerd, L.A., Hanson, W.F., Nath, R., Ibbott, G., 1999. Guidance to users of Nycomed Amersham and North American Scientific Inc., I125 interstitial sources: Dosimetry and calibration changes: Recommendations of the American Association of Physicists in Medicine Radiation Therapy Committee ad hoc Subcommittee on Low Energy Seed Dosimetry. Med. Phys. 26, 570–573.
Comprehensive Monte Carlo calculations of AAPM Task Group Report No. 43 dosimetry parameters for the Model 3500 I-Plant 125I brachytherapy source Mark J. Rivard* Department of Radiation Oncology, Tufts University School of Medicine, New England Medical Center #246, 750 Washington Street, Boston, MA 02111, USA Received 29 October 2001; received in revised form 2 January 2002; accepted 7 March 2002
Abstract The Model 3500 I-Plant 125I brachytherapy source, produced by Implant Sciences Corporation, is created through a novel technique which utilizes ion implantation of 124Xe into a ceramic matrix followed by neutron activation of the substrate. Clinical dosimetry parameters were calculated using MCNP. For data processing, a 0.376 cm active length was used for the geometry function. The 2-D anisotropy function, F ðr; yÞ; exhibited less anisotropy than the Amersham Health Model 6711 source, and decreased as radial distance, r, increased. The anisotropy constant, f% an ; was 0.933; dose rate constant, L; was 1.01770.005 cGy/h/U; and radial dose function, gðrÞ; for r={0.1, 0.2, 0.5, 2, 5, 10 cm} was 0.990, 1.014, 1.030, 0.872, 0.448, and 0.114, respectively. In general, there was good agreement with experimental results recently obtained by others using thermoluminescent dosimeters. r 2002 Elsevier Science Ltd. All rights reserved. Keywords: Monte Carlo calculations;
125
I; Brachytherapy; Dosimetry; TG-43
1. Introduction In August 2000, Implant Science Corporation (ISC) introduced the Model 3500 I-Plant 125I brachytherapy source to market. Both the design and fabrication process for this source are unique in comparison to other 125I brachytherapy sources, with the merits of this approach described by Duggan and Johnson (2001). The American Association of Physicists in Medicine (AAPM) recommends at least one complete set of experimental measurements and at least one complete set of Monte Carlo calculations preceding clinical implementation (Williamson et al., 1998). Duggan and Johnson satisfied the first recommendation obtaining measurements using thermoluminescent dosimeters (TLDs). Wallace (2002) also performed experimental measurements using TLDs, but results from Monte
*Tel.: +1-617-636-1680; fax: +1-617-636-7621. E-mail address: [email protected] (M.J. Rivard).
Carlo calculations were still needed. Therefore, the purpose of this paper is to satisfy the latter recommendation and present clinical brachytherapy dosimetry parameters as described in the AAPM Task Group Report No. 43 (TG-43) by Nath et al. (1995).
2. Materials and methods 2.1. Source geometry, composition, and energy spectrum
125
I photon
The materials and methods employed in this study were similar to those employed in previous studies by Rivard (2001a, b); the reader is referred to these papers for greater detail. A graphical depiction of the Model 3500 Monte Carlo geometry space is presented in Fig. 1. Based on physical examination of the source at various stages of fabrication, the source construction was assumed to be symmetric about the transverse-plane with the internal components (silver marker and quartz
0969-8043/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 9 6 9 - 8 0 4 3 ( 0 2 ) 0 0 1 1 0 - 0
382
M.J. Rivard / Applied Radiation and Isotopes 57 (2002) 381–389
Fig. 1. (a) Model 3500 125I source geometry as modeled using MCNP. Details of the capsule end cap weld, conical silver radiographic marker, and quartz tube are evident. The total capsule length is 4.60 mm. (b) Close-up of Model 3500 125I source geometry detailing the 124 Xe implanted quartz tube. A quartz tube is ion-implanted with 124Xe which is neutron activated to convert 124Xe into 125Xe which decays to 125I. For radiological safety purposes, the active layer is coated with a 5 mm layer of SiO2.
tube) sharing the same origin coordinates and source long-axis as the capsule. The titanium capsule inner diameter (ID) and outer diameter (OD) was 0.7239 and 0.8356 mm, respectively, with an overall length of 4.60 mm and 0.3302 mm thick hemispherical capsule end cap welds. The mass densities of the conical-ended silver marker were 10.50, 4.54 g/cm3 for surgical grade (ASTM F67-95) titanium, and 0.00120 g/cm3 for dry air modeled within the Ti capsule and used for air kerma calculations. Particular to this source is the makeup of the quartz tube (2.21 g/cm3) surrounding the 0.406 mm diameter silver marker. The quartz tube ID and OD were 0.4318 and 0.6350 mm, respectively, with a length of 3.76 mm. Atop this tube was deposited a 16 mm layer of Si into which approximately 1.5 1017 124Xe nuclei were implanted. Over this layer was deposited a 5 mm SiO2 overcoat layer to contain the radioactive 125I product. The outer diameter of the entire tube was 0.6770 mm. Based on this 124Xe loading, a mass density of 2.58 g/cm3 was calculated for the active Si/Xe layer. Following neutron irradiation and decay preceding clinical implementation, the atomic density of the active
Table 1 125 I photon source model Photon energy (keV) 27.202 27.472 30.98 31.71 35.492 28.37 keV avg.
Photons per disintegration (%) 40.6 75.7 20.2 4.39 6.68 147.6% total
Source photons with energies less than 5 keV, and source electrons, were ignored due to their relatively low yields and negligible chance of penetrating the titanium capsule.
layer was predicted to be 97.6174% Si, 2.3804% Xe, 0.0016% I, and 0.0006% Te. Using these atomic and mass densities and source geometry, the appropriate photon (MCPLIB02) and electron (EL1) cross-section libraries were chosen from the DLC-189 library supplied with the MCNP software. The 125I photon spectrum was
M.J. Rivard / Applied Radiation and Isotopes 57 (2002) 381–389
383
Table 2 Geometry function multiplied by the radial distance squared, Gðr; yÞ r2 ; for the Model 3500 I-Plant 125I brachytherapy source using the simple line equation with L ¼ 0:376 cm with Gðr0 ; y0 Þ ¼ 0:9885 r (cm)
01
101
201
301
401
501
601
701
801
901
0.25 0.50 1.00 2.00 5.00 10.00
2.3015 1.1647 1.0366 1.0089 1.0014 1.0004
2.0723 1.1558 1.0351 1.0086 1.0014 1.0003
1.6901 1.1316 1.0306 1.0075 1.0012 1.0003
1.3921 1.0978 1.0238 1.0059 1.0009 1.0002
1.1879 1.0608 1.0157 1.0040 1.0006 1.0002
1.0504 1.0258 1.0074 1.0019 1.0003 1.0001
0.9588 0.9964 0.9998 1.0000 1.0000 1.0000
0.9004 0.9745 0.9937 0.9984 0.9997 0.9999
0.8679 0.9610 0.9898 0.9974 0.9996 0.9999
0.8574 0.9565 0.9885 0.9971 0.9995 0.9999
Table 3 g(r) for the ISC I-Plant Model 3500
125
I brachytherapy source
Model 3500 r (cm) 0.050 0.075 0.100 0.150 0.174 0.200 0.250 0.275 0.300 0.350 0.400 0.450 0.500 0.600 0.700 0.750 0.800 0.900 1.000 1.250 1.500 1.750 2.000 2.250 2.500 2.750 3.000 3.500 4.000 4.500 5.000 5.500 6.000 7.000 7.500 8.000 9.000 10.000
TLDs by Duggan and Johnson (2001)
Model 6711 TLDs by Wallace (2002)
MCNP in this work
Nath et al. (1995)
1.158 1.004 0.990 1.006 1.000
1.026
1.022
1.018
1.000
1.000
0.936
0.935
0.855
0.846
0.770
0.749
0.687 0.609 0.537 0.473 0.416
0.655 0.570 0.498 0.440 0.394 0.357 0.327 0.271
0.320 0.246
0.209 0.149 0.134
1.014 1.021 1.024 1.027 1.028 1.029 1.030 1.030 1.030 1.024 1.021 1.017 1.010 1.000 0.976 0.943 0.908 0.872 0.832 0.795 0.756 0.717 0.643 0.573 0.508 0.448 0.395 0.347 0.265 0.231 0.201 0.151 0.114
1.04
1.00 0.926 0.832 0.731 0.632 0.541 0.463 0.397 0.344 0.300 0.264 0.204
Also shown are ISC Model 3500 results obtained by Duggan and Johnson (2001); Wallace (2002), and Amersham Health Model 6711 results by Nath et al. (1995).
384
M.J. Rivard / Applied Radiation and Isotopes 57 (2002) 381–389
Fig. 2. (a) Model 3500 125I gðrÞ calculated using MCNP, obtained using TLDs by Duggan and Johnson (2001) and Wallace (2002), and for the Amersham Health Model 6711 125I source as reported by Nath et al. (1995) in TG-43. (b) Close-up of the calculated Model 3500 gðrÞ data. It is evident that the dosimetry data normalized to the simple line source geometry function unexpectedly increases for r o 1 mm.The Monte Carlo-derived geometry function, or particle streaming function, permitted normalization of gðrÞ with expected results.
characterized as five photons with energies and abundances listed in Table 1, and emanated isotropically from the active layer. 2.2. Calculation techniques Monte Carlo methods using MCNP v.4B2 in a parallel-processing environment (Briesmeister, 1997;
’ yÞ in Rivard, 1999, 2000) were used to calculate Dðr; water. Dosimetry data in water were calculated at radial distances from the Ti capsule ranging from 0.050 to 10 cm, and over angles ranging from 01 to 1801 using the * F8 tally in a 30 cm diameter phantom. SK was also determined using the *F8 tally in dry air at standard temperature and pressure conditions (221C and 101.325 kPa) in a 6 m diameter phantom as a function
M.J. Rivard / Applied Radiation and Isotopes 57 (2002) 381–389
385
of distance, and by using the product of free-space air ’ free-space ; and the distance, d, squared. kerma rate, KðdÞ ’ free-space results, two corrections To obtain accurate KðdÞ were required:
3. Results and discussion
’ in-air were 1. Energy dependence of contributions to KðdÞ tracked, and contamination by K-edge titanium X’ in-air rays (4.5 and 4.9 keV) were subtracted from KðdÞ ’ in-air results were results. Specifically, the Eg¼4:7 keV KðdÞ ’ in-air subtracted from the Eg¼4:5 keV KðdÞ and ’ Eg¼4:9 keV KðdÞin-air results, these data were then ’ in-air combined and subtracted from the total KðdÞ contribution from all photons at a given distance. ’ in-air results were further corrected for 2. Corrected KðdÞ attenuation and scatter by air by extrapolating ’ free-space : ’ in-air d 2 to zero distance to obtain KðdÞ KðdÞ
The Model 3500 gðrÞ is presented in Table 3. In comparison of Model 3500 results determined using TLDs by Duggan and Johnson (2001) and Wallace (2002) in Fig. 2a, the calculated values grow larger as radius increases. This slight difference may be due to inaccurate source modeling, or may be due to errors in the media correction factor used to convert TLD results in plastic to water. For further comparison, the Model 3500 gðrÞ data are presented with the Amersham Health Model 6711 125I brachytherapy source gðrÞ data by Nath et al. (1995). The Model 3500 radial dose function appears more penetrating than the Model 6711. This increased penetration may be due to proportionally lower silver fluorescence generated in the Model 3500 source relative to the Model 6711 source. A harder photon spectrum like the 6702 is expected due to the increased spacing between the 125I active region and the silver marker due to the quartz tube. The geometric impact of the slight difference in active lengths (3.76 mm used herein vs. 4 mm used by Duggan and Johnson and also by Wallace) would become significant at the 1% and 3% levels at distances of only 3.2 and 1.4 mm along the transverse-plane, respectively. Fig 2b presents gðrÞ data at exceptionally close distances to the Ti capsule. The gðrÞ data increases beneath 1.0 mm because the line source equation no longer accurately models the photon fluence from the tubular active region. Using a Monte Carlo-derived geometry function intrinsically more representative of the photon fluence (F2 or F4), it is clear that the
These corrections are similar to those applied by Hedtj.arn et al. (2000), except here a straight-line was used for the fitting. The geometry function, Gðr; yÞ; was determined using a simple line source equation with L ¼ 0:376 cm and Gðr0 ; y0 Þ ¼ 0:9885: To ensure clarity, example values of Gðr; yÞ r2 are presented in Table 2. Gðr; yÞ was used in ’ yÞ results and data processing to normalize the Dðr; determine the remaining TG-43 parameters: % an : A total of 2 109 source L; gðrÞ; Fðr; yÞ; fan ðrÞ; and f photon histories were processed for each set of in-water and in-air calculations. Subsequent statistical uncertainties (1s) in gðrÞ and fan ðrÞ were typically less than 0.3%. Statistical uncertainties in F ðr; yÞ varied as a function of y; and were typically o 0.3% on the transverse-plane and B3% for yB01 or 1801. The statistical uncertainty in L was based on combined statistical uncertainties ’ 0 ; y0 Þ (0.07%) and KðdÞ ’ free-space from derivation of Dðr % an value (0.49%), and totaled 0.50% (1s). Finally, the f 2 was obtained by using a 1=r –weighting of fan ðr > 1 cmÞ with a 1s error likely less than 0.3%.
3.1. Radial dose function, g(r)
Table 4 ’ in-air output (MeV/g/s.p.) at various distances, relative contribution by 4.5 and 4.9 keV Ti K-edge X-rays, and Monte Carlo KðdÞ ’ in-air d 2 values corrected KðdÞ d (cm)
’ in-air MCNP output (MeV/g/s.p.) KðdÞ
Ti K-edge contribution (%)
’ in-air d 2 (MeV cm2/g/s.p.) Corrected KðdÞ
20 30 50 75 100 125 150 175 200
6.78670.041 10–7 2.98170.016 10–7 1.07370.006 10–7 4.77170.025 10–8 2.69670.014 10–8 1.72470.007 10–8 1.19170.005 10–8 8.67370.036 10–9 6.67470.020 10–9
1.6 0.82 0.32 0.13 0.064 0.031 0.015 0.009 0.006
2.672 10–4 2.661 10–4 2.673 10–4 2.680 10–4 2.694 10–4 2.693 10–4 2.679 10–4 2.656 10–4 2.670 10–4
’ free-space value, was determined by extrapolating to zero distance to remove the minor contribution of The free-space air kerma rate, KðdÞ in-air attenuation and scatter. For derivation of L, the reference dose rate (2.72070.002 10–4 MeV/g/s.p.) was divided by ’ free-space d 2 (2.67670.013 10–4 MeV cm2/g/s.p.) and found to be 1.01770.005 cGy/h/U. KðdÞ
M.J. Rivard / Applied Radiation and Isotopes 57 (2002) 381–389
386
’ 0 ; y0 Þ=ðKðdÞ ’ in-air d 2 ) values as a function of distance are presented, and are corrected by subtraction of titanium Fig. 3. Calculated Dðr K-edge X-ray contributions to in-air kerma strength. Extrapolation to zero distance permitted derivation of the Model 3500 L value1.01770.005 cGy/h/U which was in good agreement with Model 3500 TLD results (1.01 cGy/h/U) obtained by Duggan and Johnson (2001) and by Wallace (2002).
Table 5 F ðr; yÞ derived using Monte Carlo methods for the Model 3500 r (cm) 01 0.050 0.075 0.100 0.150 0.200 0.250 0.500 1.000 2.000 5.000 10.000
0.494 0.610 0.580 0.652 0.690 0.709
1
5
0.531 0.538 0.548 0.596 0.677 0.723
1
10
0.987 0.574 0.513 0.561 0.626 0.700 0.742
1
15
1.128 0.945 0.655 0.592 0.634 0.685 0.746 0.780
1
20
1.039 0.921 0.785 0.679 0.705 0.743 0.789 0.815
25
1
1.107 0.989 0.934 0.858 0.756 0.765 0.791 0.823 0.843
1
1
125
I brachytherapy source
1
30
35
40
451
501
551
601
651
701
751
801
851
1.046 0.978 0.940 0.899 0.808 0.813 0.830 0.854 0.872
1.107 1.010 0.970 0.952 0.930 0.856 0.848 0.863 0.882 0.893
1.050 0.996 0.968 0.960 0.943 0.892 0.885 0.893 0.905 0.912
1.024 0.988 0.974 0.969 0.959 0.919 0.910 0.915 0.924 0.931
1.006 0.990 0.977 0.975 0.967 0.944 0.933 0.934 0.941 0.947
0.998 0.989 0.982 0.977 0.977 0.961 0.950 0.953 0.956 0.957
1.067 0.994 0.993 0.988 0.984 0.986 0.974 0.967 0.967 0.968 0.972
1.027 0.995 0.996 0.989 0.988 0.991 0.983 0.979 0.977 0.979 0.979
0.996 0.996 0.988 0.987 0.988 0.995 0.990 0.987 0.987 0.986 0.990
0.990 0.999 1.001 0.992 0.994 0.994 0.994 0.993 0.993 0.992 0.994
0.985 0.996 0.999 0.996 0.997 1.000 0.997 0.997 0.997 0.996 0.997
0.988 1.003 1.000 0.994 0.998 0.996 1.002 1.001 1.000 0.999 1.001
By definition, F ðr; y ¼ 900 Þ was unity. There was generally greater anisotropy along the source long-axis, towards smaller y angles, and this anisotropy decreased as radial distance increased. Blank data indicate the position was located within the source where radiation dosimetry data are not clinically relevant.
supra-linear gðrÞ effect beneath 1 mm is not evident. Both data sets converge at r > 3 mm. While a more accurate equation for the geometry function was proposed by Rivard (1999), later discussion by Kouwenhoven et al. (2001) and realization of current treatment planning software limitations suggest continued use of the simple point and line equations. Therefore, the Monte Carlo-derived geometry function, with a
proposed nomenclature of the ‘‘particle streaming function’’, may be used in other applications and is not currently recommended for clinical use with conventional brachytherapy treatment planning software. For convenience of clinical implementation, the line source gðrÞ data have been fit to a fifth order polynomial with parameters given below. This equation fit the line source gðrÞ data within an absolute deviation of 71.0%
M.J. Rivard / Applied Radiation and Isotopes 57 (2002) 381–389
over a radial range of 0.1–10 cm. 2
a0 ¼ 1:005; a1 ¼ 9:422 10 ; 1 a2 ¼ 1:217 10 ; a3 ¼ 2:577 102 ; a4 ¼ 2:325 103 ; a5 ¼ 7:828 105 : 3.2. Dose rate constant In terms of Monte Carlo output, the reference dose ’ 0 ; y0 Þ; was 2.72070.002 10–4 MeV/g/s.p. rate, Dðr
387
where s.p. indicates source particle. Monte Carlo ’ in-air output at various distances, relative contribuKðdÞ tion by 4.5 and 4.9 keV Ti K-edge X-rays, and the ’ in-air d 2 values are given in Table 4. corrected KðdÞ After subtraction of titanium K-edge X-ray contribu’ 0 ; y0 Þ=KðdÞ ’ in-air d 2 ) was fit to a tions, the ratio Dðr straight line (y ¼ 1:6737 1026 w þ 1:0166) to obtain ’ free-space : Extrapolation to d ¼ 0 in Fig. 3 ’ 0 ; y0 Þ=KðdÞ Dðr provided a Model 3500 L value of 1.01770.005 cGy/h/U. Duggan and Johnson obtained 1.01 cGy/h/U, as did
Fig. 4. F ðr; yÞ derived using Monte Carlo methods for the Model 3500 125I brachytherapy source. In (b) and (c), the F ðr ¼ 1 cm; yÞ and F ðr ¼ 5 cm; yÞ data are presented, respectively, and indicate fairly good agreement with results obtained by Duggan and Johnson (2001) and by Wallace (2002) for the same type of source. In comparison, the Model 6711 source exhibits greater anisotropy.
388
M.J. Rivard / Applied Radiation and Isotopes 57 (2002) 381–389
Fig. 4 (continued).
Wallace. While Monte Carlo statistical uncertainties were relatively small, uncertainties due to source modeling, cross-section library data, low-energy photon transport physics, or errors in the plastic-to-water correction factors used with TLD results may all contribute to increase the total uncertainty. Compared with the reported Model 6702 and 6711 L values of 1.04 and 0.98 cGy/h/U presented by Williamson et al. (1999), the Model 3500 L was quite similar. This L-averaged behavior was expected since the Model 3500 produces Ag X-rays due to the silver marker (like the Model 6711), yet the 125I is in contact with a relatively low-Z backing (like the Model 6702). 3.3. Anisotropy Table 5 presents calculated Model 3500 F ðr; yÞ data. Results generally indicate that anisotropy increases as the radial distance decreases and as the angle from the transverse-plane increases. Fig. 4a graphically depicts these F ðr; yÞ data. In Figs. 4b and c, F ðr ¼ 1; yÞ and F ðr ¼ 5; yÞ are compared with result by Duggan and Johnson. General agreement is evident here, and with results obtained by Wallace with differences approaching 10% in some instances. Neither of these investigators observed the dip in F ðr; yÞ near yB100 which is likely due to capsule shielding of the source photons by the end cap weld. This is confirmed using simple geometry. Since the proportion of primary-to-scatter dose decreases as distance increases, it is not surprising that the magnitude of the 101 dip decreases as r increases. Upon comparison of Model 3500 F ðr; yÞ data with Model 6702 and 6711 F ðr; yÞ data by Nath et al.,
the Model 3500 125I brachytherapy source appears to have anisotropy more similar to the Model 6702. Table 6 presents calculated Model 3500 fan ðrÞ and fan values. Results differ by B1% with TLD measurements by Duggan and Johnson and by Wallace. The fan value did not significantly vary depending on whether 0.5 or 0.25 cm increments were used in the fan ðr > 1 cmÞ integration. As expected from the F ðr; yÞ data, Model 6711 fan ðrÞ data differ significantly from the Model 3500 fan ðrÞ data, and the Model 3500 125I source exhibits less anisotropy. Considering the combined experimental and calculative uncertainties, the calculated fan for the Model 3500 source was in good agreement with results obtained by the three previously mentioned investigators. Upon dividing each fan ðrÞ value by fan as proposed in Eq. (1) by Rivard (2001b) for a 103Pd source of significantly different construction, these Model 3500 data fit on average within 1% over the entire radial range.
4. Summary The gðrÞ; F ðr; yÞ; fan ðrÞ; and fan values obtained using Monte Carlo calculations and those obtained by others using TLDs for dosimetry of the Model 3500 125I brachytherapy source were in close agreement. Differences due to use of different active source lengths were negligble beyond 1 mm. In comparison to the Model 6711 125I brachytherapy source, the Model 3500 had more isotropic dose distributions, a slightly higher dose rate constant, and slightly increased penetration in water. There was good agreement between the
M.J. Rivard / Applied Radiation and Isotopes 57 (2002) 381–389 Table 6 1-D anisotropy functions, fan ðrÞ; and anisotropy constants, fan ; for the Model 3500 and Model 6711 125I brachytherapy sources Model 3500 r (cm)
0.25 0.30 0.35 0.40 0.45 0.50 0.60 0.70 0.80 0.90 1.00 1.25 1.50 1.75 2.00 2.50 3.00 3.50 4.00 4.50 5.00 6.00 7.00 10.00 fan
Model 6711
TLDs by TLDs by Duggan and Wallace Johnson (2002) (2001)
1.01
0.956
0.96
0.948
0.94
0.944 0.943 0.943 0.944 0.945
0.92
0.948 0.951 0.952
0.98 0.97
0.95
0.95
0.93 0.96
MCNP in this work
1.164 1.077 1.029 1.005 0.985 0.973 0.954 0.945 0.940 0.936 0.933 0.930 0.931 0.931 0.931 0.932 0.934 0.935 0.937 0.938 0.938 0.939 0.942 0.948 0.933
Nath et al.(1995)
0.944
0.936 0.893 0.887 0.884 0.880 0.901
0.93
2
fan was obtained using a 1/r –weighting for fan ðr > 1Þ data.
calculated and measured L values for the Model 3500 125 I brachytherapy source.
Acknowledgements Funding was provided through an unrestricted educational research grant provided by Implant Sciences Corporation (Wakefield, MA).
389
References Briesmeister, J.F., 1997. MCNP—A general Monte Carlo N-particle transport code system, Version 4B, LA12625-M. Duggan, D.M., Johnson, B.L., 2001. Dosimetry of the I-Plant Model 3500 iodine-125 brachytherapy source. Med. Phys. 28, 661–670. Kouwenhoven, E., van der Laars, R., Schaart, D.R., 2001. Variation in interpretation of the AAPM TG-43 geometry factor leads to unclearness in brachytherapy dosimetry. Med. Phys. 28, 1965–1966. Hedtj.arn, H., Carlsson, G.A., Williamson, J.F., 2000. Monte Carlo-aided dosimetry of the Symmetra model I25.S06 I125, interstitial brachytherapy seed. Med. Phys. 27, 1076–1084. Nath, R., Anderson, L.L., Luxton, G., Weaver, K.A., Williamson, J.F., Meigooni, A.S., 1995. Dosimetry of interstitial brachytherapy sources: Recommendations of the AAPM Radiation Therapy Committee Task Group No. 43. Med. Phys. 22, 209–234. Rivard, M.J., 1999. Refinements to the geometry factor used in the AAPM Task Group Report No. 43 necessary for brachytherapy dosimetry calculations. Med. Phys. 26, 2445–2450. Rivard, M.J., 2000. Neutron dosimetry for a general 252Cf brachytherapy source. Med. Phys. 27, 2803–2815. Rivard, M.J., 2001a. Monte Carlo calculations of AAPM Task Group Report No. 43 dosimetry parameters for the MED3631-A/M 125I source. Med. Phys. 28, 629–637. Rivard, M.J., 2001b. A discretized approach to determining TG-43 brachytherapy dosimetry parameters: Case study using Monte Carlo calculations for the MED3633 103Pd source. Appl. Radiat Isot. 55, 775–782. Wallace, R.E., 2002. Model 3500 125I brachytherapy source dosimetric characterization. Appl. Radiat. Isot. 56, 581–587. Williamson, J.F., Coursey, B.M., DeWerd, L.A., Hanson, W.F., Nath, R., 1998. Dosimetric prerequisites for routine clinical use of new low energy photon interstitial brachytherapy sources. Med. Phys. 25, 2269–2270. Williamson, J.F., Coursey, B.M., DeWerd, L.A., Hanson, W.F., Nath, R., Ibbott, G., 1999. Guidance to users of Nycomed Amersham and North American Scientific Inc., I125 interstitial sources: Dosimetry and calibration changes: Recommendations of the American Association of Physicists in Medicine Radiation Therapy Committee ad hoc Subcommittee on Low Energy Seed Dosimetry. Med. Phys. 26, 570–573.