Ponente
Sr.
Álvaro Perales
(Universidad de Sevilla)
Descripción
Introduction
The modeling of tissue heterogeneities and photon fluences for complex treatment plans in radiotherapy has been reported as the main source of discrepancy in dose calculation [1]. Dose distribution disagreements between Monte Carlo (MC) and treatment planning system (TPS) calculations produce significant variations in the estimation of tumor control probability (TCP) [2]. Hence for a complete treatment planification including radiobiological parameters we have to obtain an accurate calculation of the dose distribution.
Our purpose is to assess radiobiological variations of lung cancer treatment plans due to divergences between dose distributions calculated by Pinnacle³ TPS (version 9.8), employed at the Hospital Universitario Virgen Macarena (Seville, Spain), and by the Geant4 MC toolkit (version 10.01.p01) [3,4].
Material and methods
Three clinical cases (see Table 1) distinguished by their tumoral size and localization were analysed. Each clinical case presents a treatment planification composed by 6 MV photon beams with a conventional fractionation, i. e. 2 Gy per fraction. This scheme was established in order to achieve a 66 Gy prescribed dose to the 95% of the planning target volume (PTV). We have set conformity index (CI) and homogeneity index (HI) [5] as key parameters for dose distribution comparisons between Pinnacle3 and Geant4.
In addition to this dose analysis, we have calculated TCP according to the lineal-quadratic (LQ) model [6]. For this task we have used an alpha value of 0.3 Gy-1, a beta value of 0.03 Gy-2 and a density of clonogenic cells of 10^7 cells/cm3.
For TPS calculations we have applied the Collapsed Cone algorithm and a voxel resolution of 2 x 2 x 2 mm³. In our MC calculations we have simulated 2•10^10 events per clinical case. The scoring volume dimensions were equal to patient DICOM image files resolution, i. e. 0.937 x 0.937 x 5 mm³. These files were exported from the TPS for every single treatment plan.
Results
In Table 2 we have summarized the CI, HI and TCP values calculated with Geant4 and Pinnacle3 for each clinical case. Furthermore, we have obtained variations of absorbed dose at the 95% of the PTV (between Geant4 and TPS dose calculations) up to 2.3 Gy for the clinical case 1. In this case, a gamma analysis [7] between calculated dose distributions with Geant4 and TPS (Fig. 1), done with acceptance criteria of 3%-3mm, gave a passing rate of 78.3% within the patient. This evaluation was done utilizing Monte Carlo Treatment Planning (MCTP) CARMEN [8,9].
Discusion and conclusions
In all clinical cases we have found more heterogeneous dose distributions employing Geant4 as calculation system. HI and CI values show us that dose calculations depend on the particular clinical case. For clinical case 1 we have done a gamma analysis (3%-3 mm), which shows high discrepancies (red points which do not satisfied gamma criteria) at the dose maximum zone which is inside PTV. Dose divergences, expressed through HI and CI values, gave variations in TCP close to 4% between Geant4 and TPS calculations. In our methodology we have used the LQ model with an alpha/beta ratio of 10 Gy and a constant clonogenic cells density of 10^7 cells/cm3. A further study including more sophisticated radiobiology models and hypofractionated radiotherapy schemes will is ongoing.
Acknowledgments
This work was funded in part by the Spanish Ministry of Economy and Competitiveness (under projects no. FPA2014-53290-C2-2-P and FPA2016-77689-C2-1-R) and by Junta de Andalucía (under project no. P12-FQM-1605). The MC simulations were performed with our FIS-ATOM computing cluster, hosted at Centro Informático y Científico de Andalucía (CICA, Seville, Spain).
References
[1] B. Vanderstraeten, et al. Medical Physics 33 (2006) 3149.
[2] I. J. Chetty et al. Radiother. Oncol. 109 (2013) 498.
[3] S. Agostinelli et al. Nucl. Instrum. Meth. A 506 (2003) 250.
[4] J. Allison et al. Nucl. Instrum. Meth. A 835 (2016) 225.
[5] ICRU Report 83. Prescribing, recording, and reporting intensity-modulated photon-beam therapy (IMRT). 2010.
[6] J. F. Fowler Br. J. Radiol. 62 (1989) 679.
[7] D. Low et al. Medical Physics 25 (1998) 656.
[8] A. Ureba et al. Medical Physics 41 (2014) 081719-1.
[9] J. A. Baeza et al. Medical Physics 42 (2015) SU-E-T-157.
Autor primario
Sr.
Álvaro Perales
(Universidad de Sevilla)
Coautores
Prof.
María Isabel Gallardo Fuentes
(Departamento de Física Atómica, Molecular y Nuclear. Universidad de Sevilla)
Dr.
Miguel Antonio Cortes-Giraldo
(Universidad de Sevilla)
