A modern Monte Carlo investigation of the TG-43 dosimetry parameters for an 125I seed already having AAPM consensus...

To investigate potential causes for differences in TG-43 brachytherapy dosimetry parameters in the existent literature for the model IAI-125A(125)I seed and to propose new standard dosimetry parameters.

The MCNP5 code was used for Monte Carlo (MC) simulations. Sensitivity of dose distributions, and subsequently TG-43 dosimetry parameters, was explored to reproduce historical methods upon which American Association of Physicists in Medicine (AAPM) consensus data are based. Twelve simulation conditions varying(125)I coating thickness, coating mass density, photon interaction cross-section library, and photon emission spectrum were examined.

Varying(125)I coating thickness, coating mass density, photon cross-section library, and photon emission spectrum for the model IAI-125A seed changed the dose-rate constant by up to 0.9%, about 1%, about 3%, and 3%, respectively, in comparison to the proposed standard value of 0.922 cGy h(-1) U(-1). The dose-rate constant values by Solberg et al. ["Dosimetric parameters of three new solid core (125)I brachytherapy sources," J. Appl. Clin. Med. Phys. 3, 119-134 (2002)], Meigooni et al. ["Experimental and theoretical determination of dosimetric characteristics of IsoAid ADVANTAGE™ (125)I brachytherapy source," Med. Phys. 29, 2152-2158 (2002)], and Taylor and Rogers ["An EGSnrc Monte Carlo-calculated database of TG-43 parameters," Med. Phys. 35, 4228-4241 (2008)] for the model IAI-125A seed and Kennedy et al. ["Experimental and Monte Carlo determination of the TG-43 dosimetric parameters for the model 9011 THINSeed™ brachytherapy source," Med. Phys. 37, 1681-1688 (2010)] for the model 6711 seed were +4.3% (0.962 cGy h(-1) U(-1)), +6.2% (0.98 cGy h(-1) U(-1)), +0.3% (0.925 cGy h(-1) U(-1)), and -0.2% (0.921 cGy h(-1) U(-1)), respectively, in comparison to the proposed standard value. Differences in the radial dose functions between the current study and both Solberg et al. and Meigooni et al. were <10% for r ≤ 5 cm, and increased for r > 5 cm with a maximum difference of 29% at r = 9 cm. In comparison to Taylor and Rogers, these differences were lower (maximum of 2% at r = 9 cm). For the similarly designed model 6711 (125)I seed, differences did not exceed 0.5% for 0.5 ≤ r ≤ 10 cm. Radial dose function values varied by 1% as coating thickness and coating density were changed. Varying the cross-section library and source spectrum altered the radial dose function by 25% and 12%, respectively, but these differences occurred at r = 10 cm where the dose rates were very low. The 2D anisotropy function results were most similar to those of Solberg et al. and most different to those of Meigooni et al. The observed order of simulation condition variables from most to least important for influencing the 2D anisotropy function was spectrum, coating thickness, coating density, and cross-section library.

Several MC radiation transport codes are available for calculation of the TG-43 dosimetry parameters for brachytherapy seeds. The physics models in these codes and their related cross-section libraries have been updated and improved since publication of the 2007 AAPM TG-43U1S1 report. Results using modern data indicated statistically significant differences in these dosimetry parameters in comparison to data recommended in the TG-43U1S1 report. Therefore, it seems that professional societies such as the AAPM should consider reevaluating the consensus data for this and others seeds and establishing a process of regular evaluations in which consensus data are based upon methods that remain state-of-the-art.