Monte Carlo simulation and dosimetric studies for small fields in lung stereotactic body radiation therapy

• Main Author: Soh, Roger Cai Xiang
• Other Authors: Lew Wen Siang
• Format: Theses and Dissertations
• Language: English
• Published: 2018
• Subjects: DRNTU::Science::Physics::Radiation physics
• Online Access: http://hdl.handle.net/10356/73426
• Institution: Nanyang Technological University

The use of small fields in radiotherapy has increased over the recent years as complex and precise radiotherapy techniques are being developed. However, prediction and verification of small field doses continue to prove to be difficult due to several chal- lenges such as lateral electron disequilibrium (LED), partial source occlusion effect and detector perturbation. Accurate prediction of small field doses can be achieved using Monte Carlo (MC) simulations. MC simulations are extremely versatile and can be used to accurately predict doses in any medium provided that the linear accelerator (linac) treatment head and associated parameters are well modelled. The aim of this thesis is to use MC simulation to study the accuracy of commercial dose calculation algorithms used in treatment planning and to correct dose perturbation for radiation dosimeters for small lung SBRT fields. This is done using a three pronged approach, (1) by the determination of optimal small field parameters for MC simulation (2) by evaluating the accuracy of dose calculation algorithms for small lung SBRT field sizes in the clinical treatment planning system (TPS) and (3) by accurate measurement of dose profiles using calibrated film and thermoluminescence small field dosimetry. The first step for MC simulation is to generate a clinically realistic MC linac model by identifying the optimal parameters for accurate MC simulation. Electron Gamma Shower Monte Carlo Simulation Package developed by the Ionizing Radiation Standards Group at National Research Council of Canada (EGSnrc) will be used as the MC code. Field-size specific MC simulated doses of a 6 MV photon beam were validated using 1D gamma analysis. Simulated data were benchmarked against measurements obtained using EDGE Detector (Sun Nuclear Corporation, Melbourne, FL) and Diode SRS detector (PTW-60018, Freiburg, Germany) in a 3D scanning water tank (Sun Nuclear Corporation, Melbourne, FL). Initial MC reference parameters were approximated using average percentage depth and lateral dose profiles differences as simulated from different mean electron energy and electron beam radial distribution (full width at half maximum, FWHM). Subsequently, the optimal parameters were obtained by 1D gamma analysis using increasingly stringent gamma criteria from γ0.3%/0.3mm to γ2.0%/2.0mm for depth dose and lateral dose profiles. By using a 95% passing rate, a generic set of optimal primary electron beam parameters in a MC model for all field sizes was accurately determined. The simulated 6 MV photon beam MC model was used as a benchmark to evaluate two versions of two dose calculation algorithms available in the Eclipse TPS (Varian Medical System Inc, USA), namely, Anisotropic Analytical Algorithm (AAA) version 10.0 (AAAv10.0), AAA version 13.6 (AAAv13.6) and Acuros XB dose calculation (AXB) algorithm version 10.0 (AXBv10.0), AXB version 13.6 (AXBv13.6). Depth dose distributions for 6 MV flattening filter-free (6X FFF) photon beam were also included. The study was done on a respiratory motion phantom (QUASAR, Modus Medical Devices, London, ON) with a moving chest wall. For the chest wall, a 4 cm thick wax mould was attached to the lung insert of the phantom. Average intensity projection (AIP) images from the four-dimensional computed tomography (4D-CT) scans were used for depth dose calculation. Depth doses along the central axis were compared in the anterior and lateral beam direction for field sizes 2 × 2 cm2, 4 × 4 cm2 and 10 × 10 cm2. For the lateral beam, the surface dose of the moving chest wall highlighted differences of up to 105% for AAAv10.0, 40% for AXBv10.0 and 20% for 6X FFF from MC simulations. AAAv13.6 and AXBv13.6 agrees with MC predictions to within 10% at similar depth. For anterior beam doses, dose differences for both versions of AAA, AXB algorithm and 6X FFF beam were within 7% and results were consistent with static heterogeneous studies. Small field dosimetry within a lung phantom can be a challenge due to dose pertur- bation of the detector within the low-density lung medium. For thermoluminescence dosimeters (TLDs), accurate measurement can be achieved by correcting the perturbed dose using TLD correction factors calculated from MC simulations. MC calculated TLD correction factors were compared against depth and field size within a static lung phantom. It was found that the greatest TLD correction needed was for the smallest field size investigated, 2×2 cm2, where LED was the greatest. Correction at interfaces of cork (lung substitute medium) and water were dependent on the medium before the interface. In addition, the MC corrected TLD measurement was compared to AXBv13.6 2 × 2 cm2 dose calculation to test the accuracy of the calculated correction factors. Finally, TLD corrected measurements and radiochromic films (Gafchromic External Beam Therapy 3 (EBT3), International Specialty Products, Wayne, NJ) were used to verify the MC and TPS simulated depth doses in the respiratory lung motion phantom. Both MC and AXB v13.6 depth doses were found to have the best agree- ment to within 2% deviation. A VMAT lung SBRT treatment plan with three arcs was computed with AXBv13.6, AAAv12.6, AXBv10.0 and AAAv10.0 The deliverability of the treatment plan delivery was verified using a diode array (ArcCHECK, Sun Nuclear Corporation, Melbourne, USA). Quality assurance (QA) of the plan was alsodone using film dosimetry. 2D gamma analysis show that AXB v13.6 have the best agreement with film dosimetry and diode array measurements. In conclusion, MC simulation model developed in this thesis was used to validate the accuracy of commercial dose calculation algorithms and determine the corrections needed for TLD in small fields and lung medium. The MC simulation: (1) when correctly modelled and validated by measurement, can be used as an independent benchmark for studying the accuracy commercial treatment planning systems; (2) is suitable in correcting TLD small field measurements; (3) is versatile, allowing users to verify the accuracy of treatment delivery even in extreme inhomogeneous conditions. These are important findings that will contribute to small photon field dosimetry for safer lung SBRT treatments.