Monte Carlo dosimetric study of the Flexisource Co-60 high dose rate source

Purpose

According to the American Association of Physicist in Medicine (AAPM) and the European Society for Radiotherapy and Oncology (ESTRO) recommendations [1], in order to fulfil the dosimetric prerequisites, all sources to be used in clinical practice have to have a set of dosimet­ric para­meters available based on the Radiation Therapy Committee Task Group No. 43 (TG-43) formalism [2,3].

The AAPM High Energy Brachytherapy Source Dosimetry Working Group (HEBD-WG) recommends [1] that this data­set must be based upon at least one experimental study and at least one Monte Carlo (MC) study of the model’s source dosimetric parameters. For conventionally encapsulated sources similar in design to existing or previously existing ones, a single dosimetric study published in a peer-reviewed journal is sufficient. The high dose rate (HDR) 60Co sources fall in this category. MC or experimental dosimetry (or both) me­thods may be used. These studies must be performed by investigators that are independent from the manufacturer and published in a peer-reviewed journal prior to the use of these isotopes in clinical practice.

The HEBD-WG is also concerned about the dosimetry in the near-source region where the influence of  electrons and the lack of electronic equilibrium are frequently neglected [4-6]. Commercial treatment planning systems (TPS) allow direct introduction of tabulated dose rates from the literature using the TG-43 formalism. These TG-43 data are usually derived from MC radiation transport simulations, estimating absorbed dose by collisional kerma.

Consequently, these data are provided at distances from the source capsule large enough to ensure that the equivalence of collisional kerma and dose is applicable. TPS extrapolate data outside the available TG-43 data range.

In case of HDR 60Co sources, kerma to dose differences are significant and source model specific [4]. Kerma extrapolation at short distances would not be necessary if TG-43 data were available with adequate range and spatial resolution that include source electron contributions to absorbed dose and account for electron disequilibrium.

High dose rate (HDR) brachytherapy 60Co sources have been considered for use in clinical practice as an alternative for 192Ir HDR sources [7]. A comparison between the radio­logical properties of cobalt and iridium HDR sources have been performed in reference [8]. Additionally to cost and logistics improvements due to the cobalt longer half life, clinical examples for intracavitary and interstitial applications show practically identical dose distributions.

The main goal of this work is to present the TG-43 data and the 2D dose rate table in cylindrical coordinates for treatment planning and quality assurance purposes (QA) for the new Flexisource Co-60 HDR source model used by the Flexitron remote HDR afterloader (Nucletron B.V., Veenendaal, The Netherlands) in a consistent way that is valid at short and long distances. Such a source has not been studied and published previously.

Material and methods

The design and materials of the Flexisource Co-60 HDR source was provided by the manufacturer. The source design and dimensions are shown schematically in Fig. 1. It is composed of a central cylindrical active core made of metallic 60Co with a density of 8.9 g/cm³, 3.5 mm in length and 0.5 mm in diameter. The active core is covered by a cylindrical 316L stainless steel (67% Fe, 11% Ni, 18% Cr, 2% Si, 2% Mn) layer of 0.9 mm of external diameter and a density of 8 g/cm³. For this study we considered 2 mm 316L steel cable with an effective density of 4.81 g/cm³ (measured from the inner clamp). The interstitial areas between the active element and the cover were considered to be filled with standard dry air.

MC methods for radiation transport simulations were used to study the dose rate distribution around the source. Different MC codes presented different physics models, different cross sections data, and dissimilar tracking me­thods in the transport of electrons. As dose at short distances from the source was required, where electronic disequilibrium conditions may be dosimetrically important, two different MC codes were used. These MC codes were Penelope2008 [9] and GEANT4 (version 9.3) [10], which have been successfully used for dosimetric studies in the field of the bra­chytherapy [4,11-15].

GEANT4 and Penelope2008 photon and electron cross-sections are based on the EPDL97 and EEDL97 cross sections libraries, respectively [16, 17]. However, Penelope2008 also considers the impulse approximation that accounts for Doppler broadening and binding effects [9]. Consequently, photoelectric effect, pair production and Rayleigh cross-sections used by both codes were the same, while Compton cross-sections in Penelope2008 differed from those of GEANT4. Possible influence on the dosimetric results by using Penelope2008 with the Compton cross-sections of the EPDL97 library has been discussed elsewhere [14] and found to be negligible. The photon spectrum was taken from the NuDat database [18] as suggested in Ref. [19]. The number of photons N generated in each simulation was as follows: Penelope2008 (N = 5 × 109), GEANT4 (N = 1 × 109 to obtain water kerma and N = 4 × 109 to calculate absorb­ed dose). The electron spectrum including  decay, internal conversion electrons (IC) and Auger electrons was not considered in the simulation since its effect over the total dose was known to be less than 1% at distances greater than 0.1 cm from the source surface [4]. In each disintegration, 2.0001 photons/(Bq s) were generated on average. However, due to the 10 keV cut-off used, the photon intensity was reduced to 1.9985 photons/(Bq s).

Dose to water contribution was calculated in Penelope2008 using the tally provided within the penEasy package [20], whereas in GEANT4 it was evaluated using a homemade routine with the function GetTotalEnergyDeposit of the GEANT4 toolkit. To estimate the collisional kerma, homemade routines using the linear track-length estimator [21] were developed for Penelope2008 and GEANT4. Dose and collisional kerma rate distributions were used to derive the final dosimetric parameters as a function of r at every polar angle sampled.

The source was located at the geometric center of a sphe­rical liquid water phantom with 100 cm radius, to estimate dose to water and simulate unbounded phantom conditions for r < 20 cm [22]. Water composition and mass density were those recommended by the AAPM [3]. Due to the high energy of the 60Co, the photon spectrum electronic equilibrium is not reached up to a distance of approximately 0.75 cm from the surface of the source [4]. Thus, the dose for small distances cannot be approximated by collisional kerma as is usually done for 192Ir or 137Cs sources. Differences between collisional kerma and dose at r = 0.75 cm are less than 0.5% and negligible at r = 1 cm [4]. Since the evaluation of collisional kerma was more efficient (reduced statistical uncertainty and improved numerical performance) we have considered absorbed dose to water for distances smaller than 0.75 cm and collisional kerma from 0.75 cm up to 20 cm. In order to provide adequate spatial resolution, the cells were 0.01 cm in thickness for r < 2 cm from the source and a factor of 10 thicker for 2 cm < r < 20 cm, respe­ctively. Collisional kerma and absorbed dose were obtained simultaneously in cylindrical (y,z) and spherical (r,) coordinates. Angular sampling was taken every 2°.

Additional simulations were performed to obtain SK with the source surrounded by vacuum, except for a small cylindrical air cell of 0.1 cm in diameter and 0.1 cm in height at r = 10 cm, as recommended by AAPM [3]. Mass-energy absorption coefficients in water and air were consistently derived for each code and used to calculate the collisional kerma.

Results

The dose rate distribution D.(r,) for the Flexisource

Co-60 HDR source model constructed as described in Sect. Material and Methods was used to derive the TG-43 do­simetry parameters with L = 0.35 cm. Using Penelope2008 and GEANT4, an average of  = 1.085 ± 0.003 cGy/(h U) (with k = 1) was obtained. Uncertainties are Type A only. These are similar to the consensus values published for other 60Co sources, see Table 1. In Table 2 and 3 gL(r) and F(r,) are provided in 0.1 cm steps (or smaller) up to 1.0 cm from the source (to reproduce the dose distribution accurately at close distances) and in 0.5 cm and 1 cm steps up to 20 cm. Both functions were obtained as average results from Penelope2008 and GEANT4 codes. F(r,θ) was provided for all radial distances in 2° increments. An along-away table for QA purposes is also provided in Table 4.

Differences in using D(r,) Penelope2008 and GEANT4 were within the statistical uncertainties (type A). These uncertainties were larger at r < 1 cm where absorbed dose were scored (between 0.5% at r = 0.2 cm and 1% at r = 0.8 cm) and lower at larger distances (below 0.1%) where the collisional kerma was used.

Discussions

In this study, we have compared our results with those obtained for other 60Co sources discussed in the literature. Papagiannis et al. [23] used MC to obtain dose rate in water (30 cm in diameter water phantom) of the Ralstron Type-1, Type-2 and Type-3 source models manufacured by Shimazdu Corporation (Japan) and used in the Ralstron remote afterloader. Their configuration consists of two active pellets (cylinders 1 mm × 1 mm) either in contact (Type-2), 9 mm (Type-1) or 11 mm apart (Type-3). All three models have a 3 mm external diameter. Kerma-dose approximation was used. Papagiannis et al. [23] reported along-away dose rate tables and TG-43 dose parameters. Selvam et al. [24] have reported a systematic error for y = 0.75 cm in the away-along table of Papagiannis et al. for the type 2 source model.

Ballester et al. [13] studied the GK60M21 60Co (Eckert & Ziegler IBt-Bebig GmbH, Germany) using GEANT4 code to obtain the dose rate distribution around this source in an unbounded water phantom. Only the gamma part of the 60Co spectrum was considered. The  spectrum contribution to the dose was assumed to be insignificant. A cut-off energy of 10 keV was used for both photons and electrons. They scored kerma and dose separately to account for the electronic disequilibrium near the source. For points located at distances of less than 1 cm from the source they scored dose, while for distances where electronic equilibrium was achieved they scored kerma. They derived TG-43U1 parameters and an away-along table. Selvam et al. [25] repro­duced the Ballester et al. [13] study, but using the EGSnrc code. They derived only an away-along table. The comparison of away-along tables from both studies reveals consistency between both studies except at y = 0.25 cm and z = –0.25, z = 0 and z = 0.25 cm were the Ballester et al. data had a typo.

Granero et al. [11] used GEANT4 MC code to obtain the dose rate distribution for the Co0.A86 60Co source model (Eckert & Ziegler IBt-Bebig GmbH, Germany) in an unbound­­ed water phantom. The same type of study that the Balle­ster et al. [13], one of the GK60M21 source model described in the previous paragraph was done for the Co0.A86 source model. Selvam et al. [25] also reproduced the Granero et al. [11] study, but using the EGSnrc code, obtaining only an away-along table. The comparison of away-along tables from both studies reveals that at (y = 0.25 cm, z = –0.25 cm), (y = 0.25 cm, z = 0 cm), (y = 0.25 cm, z = 0.25 cm), the Granero et al. [11] data are underestimated. This is the same typo as in Ballester et al. data [13] for the GK60M21 source model.

Tabrizi et al. [26] studied two different 60Co linear braid type sources available for the GZP6 remote afterloader (Nuclear Power Institute of China). These sources are composed of one active core made of metallic 60Co with 3.5 mm length and 1.5 mm diameter, encapsulated in 0.1 mm titanium.

The active core is covered by a cylindrical stainless steel cover of 0.5 mm external diameter and steel balls arranged along. The authors used the MCNP4C Monte Carlo code to obtain the TG-43 dosimetric parameters together with along-away dose rate tables. Their radial dose function (see Fig. 2 in Tabrizi et al.) is inconsistent with other 60Co source data and is difficult to understand from a physical point of view.

Papagianis et al. [23] showed that /G (r = 1 cm,  = 90°) values for different 60Co source models are expected to match each other providing the spatial dependence of the dose rate constant as removed. In Table 1, it can be observed that the Flexisource Co-60 HDR source fits into this scheme.

gL(r) for the Flexisource Co-60 HDR source is compared in Fig. 2A with corresponding data to GK60M21 [13] and Co0.A86 [11] sources from BEBIG, Ralstron HDR Type 2 [23] from Shimazdu and GZP6 sources [26]. Figure 2B illustrates similarities/differences for r < 1 cm. The differences between Papagianis et al. data and those of the present study are due to the different phantom sizes used in the MC calculations. For the GZP6 source model data present an anomalous pattern.

Anisotropy function F(r,) is shown in Fig. 3 for r  1 cm. At larger distances, F(r,) behave in a similar way as for r = 1 cm.

Conclusions

A dosimetric study of the Nucletron Flexisource Co-60 HDR source for which no published dosimetric data existed was performed. TG-43 parameters, dose rate constant, radial dose function and anisotropy function were provided together with a 2D along and away dose table. These datasets can be used either as an input for (or to validate) the TPS calculations essential for clinical practice.

Acknowledgements

This study was supported in part by Generalitat Valen­ciana (Project PROMETEO2008/114) FEDER, Ministerio de Ciencia e Innovación, Spain (Project No. FIS2010-17007) and by a research agreement with Nucletron B.V. Veenendaal, The Netherlands.

References

 1. Li Z, Das RK, DeWerd LA et al.; American Association of Physicists in Medicine (AAPM); European Society for Therapeutic Radiology and Oncology (ESTRO). Dosimetric prere-quisites for routine clinical use of photon emitting brachytherapy sources with average energy higher than 50 keV. Med Phys 2007; 34: 37-40.

 2. Nath R, Anderson LL, Luxton G. et al. Dosimetry of interstitial brachytherapy sources: recommendations of the AAPM Radiation Therapy Committee Task Group No. 43. American Association of Physicists in Medicine. Med Phys 1995; 22: 209-234.

 3. Rivard MJ, Coursey BM, DeWerd LA et al. Update of AAPM Task Group No. 43 Report: A revised AAPM protocol for brachytherapy dose calculations. Med Phys 2004; 31: 633-674.

 4. Ballester F, Granero D, Pérez-Calatayud J et al. Evaluation of high-energy brachytherapy source electronic disequilibrium and dose from emitted electrons. Med Phys 2009; 36: 4250-4256.

 5. Baltas D, Karaiskos P, Papagiannis P et al. Beta versus gamma dosimetry close to Ir-192 brachytherapy sources. Med Phys 2001; 28: 1875-1882.

 6. Wang R, XA Li. A Monte Carlo calculation of dosimetric para­meters of 90Sr/90Y and 192Ir SS sources for intravascular bra­chytherapy. Med Phys 2000; 27: 2528-2535.

 7. Baltas D, Lymperopoulou G, Zamboglou N. On the use of HDR 60Co source with the MammoSite radiation therapy system. Med Phys 2008; 35: 5263-5268.

 8. Richter J, Baier K, Flentje M. Comparison of 60-cobalt and 192-iridium sources in high dose rate afterloading brachytherapy. Strahlenther Onkol 2008; 184: 187-192.

 9. Salvat F, Fernández-Varea JM, Sempau J. PENELOPE-2008: a code system for Monte Carlo simulation of electron and photon transport, 2008. Issy-les-Moulineaux, France.

10. GEANT4 Collaboration. GEANT4 – a simulation toolkit. Nucl Instrum Methohs Phys Res 2003; 506: 250-303.

11. Granero D, Perez-Calatayud J, Ballester F. Technical note: Dosimetric study of a new Co-60 source used in brachytherapy. Med Phys 2007; 34: 3485-3488.

12. Granero D, Perez-Calatayud J, Pujades-Claumarchirant MC et al. Equivalent phantom sizes and shapes for brachytherapy dosimetric studies of 192Ir and 137Cs. Med Phys 2008; 35: 4872-4877.

13. Ballester F, Granero D, Pérez-Calatayud J et al. Monte Carlo do­simetric study of the BEBIG Co-60 HDR source. Phys Med Biol 2005; 50: N309-N316.

14. Granero D, Vijande J, Ballester F et al. Dosimetry revisited for the HDR 192Ir brachytherapy source model mHDR-v2. Med Phys 2011; 38: 487-494.

15. Meigooni AS, Wright C, Koona RA et al. TG-43 U1 based dosimetric characterization of model 67-6520 Cs-137 brachytherapy source. Med Phys 2009; 36: 4711-4719.

16. Cullen DE, Hubbell JH, Kissel L. EPDL97: the Evaluated Photon Data Library, ‘97 Version, 1997, University of California, Lawrence Livermore National Laboratory: Livermore, CA.

17. Cullen DE, Perkins ST, Seltzer SM. Tables and graphs of electron-interaction cross-sections from 10 eV to 100 GeV derived from the LLNL Evaluated Electron Data Library (EEDL) Z=1–100, 2001, Lawrence Livermore National Laboratory.

18. NUDAT 2.5, National Nuclear Data Center, 2010, Brookhaven National Laboratory.

19. Rivard MJ, Granero D, Perez-Calatayud J et al. Influence of photon energy spectra from brachytherapy sources on Monte Carlo simulations of kerma and dose rates in water and air. Med Phys 2010; 37: 869-876.

20. Badal A, Sempau J. A package of Linux scripts for the parallelization of Monte Carlo simulations. Comput Phys Commun 2006; 175: 10.

21. Williamson JF. Monte Carlo evaluation of kerma at a point for photon transport problems. Med Phys 1987; 14: 567-576.

22. Perez-Calatayud J, Granero D, Ballester F. Phantom size in brachytherapy source dosimetric studies. Med Phys 2004; 31: 2075-2081.

23. Papagiannis P, Angelopoulos A, Pantelis E et al. Monte Carlo dosimetry of 60Co HDR brachytherapy sources. Med Phys 2003; 30: 712-721.

24. Selvam TP, Sahoo S, Vishwakarma RS. Comment on “Monte Car­lo dosimetry of 60Co HDR brachytherapy sources” [Med Phys 30, 712-721 (2003)]. Med Phys 2010; 37: 5146-5147.

25. Selvam TP, Bhola S. Technical note: EGSnrc-based dosimetric study of the BEBIG 60Co HDR brachytherapy sources. Med Phys 2010; 37: 1365-1370.

26. Hariri Tabrizi S, Kamali Asl A, Azma Z. Monte Carlo derivation of AAPM TG-43 dosimetric parameters for GZP6 Co-60 HDR sources. Phys Med 2012; 28: 153-160.