ESTRO 37 Abstract book

S939

ESTRO 37

measurement should also be carried out until confidence in the MPC is achieved. EP-1750 Secondary Dose Calculation: Detecting Systematic Errors in the Treatment Planning System Beam Model J. Wong 1 , W. Warren 1 , K. Homann 1 1 Houston Methodist Hospital, Radiation Oncology, Houston, USA Purpose or Objective The purpose of this study was to evaluate a commercial secondary dose calculation software’s (SDCS) ability to detect systematic errors in a treatment planning system (TPS) beam model for model validation and treatment plan approval. Material and Methods Artificial changes to the TPS beam model introduced errors to patient plans. 12 cases were analyzed with the SDCS to determine effects of the following: 1. Incorrect TPS settings for leaf-end modeling regarding (a) MLC offset table, and (b) leaf tip radius, 2. Incorrect TPS settings to account for tongue-and-groove effects, 3. Dose gradient errors due to inaccurate volume-averaged dose profiles, 4. Inherent dose gradient errors in the TPS beam model/algorithm, and 5. TPS underestimation of dose for narrow MLC segments. Results Plans were evaluated with the SDCS for gamma dose criteria, mean target dose, dose coverage, dose volume histogram (DVH), and profiles. Study 1(a) resulted in no plans failing when adjustments to the MLC offset table in the TPS beam model were made. However, a decrease in the mean target dose and 90% dose coverage can been seen throughout all plans. Study 1(b) showed 10 plans failing at 2%/2mm when the leaf tip radius was increased by 6 cm. In an example case study, a prostate VMAT passed at 99.2% and decreased to 98.7% at 3%/3mm when the leaf tip radius increased from 10 cm to 16 cm. At 2%/2mm the pass rate finally failed at 89.4% for a 16 cm leaf tip radius (Figure 1). Study 2 showed minimal differences in passing rates when not accounting for the tongue-and-groove effects. However, an increase in the mean target dose and 90% dose coverage is seen for all but one plan. Differences in DVH are also seen, specifically in the example of a lung SBRT at 2%/2mm, where the pass rate was 97.7% for the original plan and 97.2% when the beam model was adjusted. The SDCS target dose was higher at 68.9 Gy as compared to the calculated target dose of 68.5 Gy for the original plan. The SDCS DVH curve shifted about 3.8% at D95 (Figure 2). Study 3 showed 2 plans failing at 2%/2mm and 10 plans with decreased gamma values when the profile data from a CC13 chamber replaced data taken with a CC01. Study 4 reviewed the original plans which revealed an underlying dose pattern corresponding to overestimation and underestimation of dose throughout. Inspection of the dose profiles showed clear systematic dose gradient errors. The study also revealed that as the passing criteria becomes tighter or as the dose grid becomes smaller, the pass rates decrease. Study 5 further attributed dose differences to narrow MLC segments, especially in highly-modulated cases. Individual dose profiles showed data consistent with reduction of dose in the high dose region and overestimation of dose for narrow segments.

Conclusion This study verified errors in commissioning measurement and beam modeling can be detected with the SDCS. However, large modeling errors or tighter tolerances are needed before clinical failures can be detected. EP-1751 Accuracy of Monte Carlo software PRIMO against a reference dosimetry dataset for 6 MV photon beams M. Hermida-López 1 , D. Sánchez-Artuñedo 1 , J.F. Calvo- Ortega 2 1 Hospital Universitario Vall d'Hebron, Servicio de Física y Protección Radiológica, Barcelona, Spain 2 Hospital Quirón, Servicio de Radioterapia, Barcelona, Spain Purpose or Objective The parameters that characterize the initial electron beam in the simulation of a clinical linac greatly influence on the simulated dose distributions. The users of a Monte Carlo code for the simulation of radiation transport must match the simulated dose distributions to measurements and this is achieved usually in a lengthy iterative process, by tuning the simulation parameters and comparing the simulated with the measured dose distributions. To reduce the time needed for this tuning task, the PRIMO Monte Carlo software for the simulation of clinical linacs [Strahlenther. Onkol. 189 , 881–886 (2013)] proposes default values of the initial beam parameters for each nominal energy of the included linac models. This work investigates the suitability of the PRIMO default beam parameters for the 6 MV photon beam from Varian Clinac 2100 linacs with a Millennium 120 MLC, by comparing dosimetric data obtained from PRIMO simulations with a published dataset based on measurements on a large series of linacs of the same model. Material and Methods We used PRIMO v. 0.3.1.1363 to obtain a phase-space file (PSF) with 450 million histories of a Clinac 2100 with a 6 MV photon beam (E=5.4 MeV by default in PRIMO). We used this PSF to estimate dosimetric parameters in a water phantom for some setups of interest (static fields). Point measurement distributions provided by the Imaging and Radiation Oncology Core-Houston (IROC-H) Quality Assurance Center were used as benchmark data [Med. Phys. 43 (5), 2374–2386 (2016)]. The accuracy of PRIMO simulation results was assessed for percentage depth doses, jaw output factors, MLC small-field output factors, and off-axis ratios. Results All evaluated dosimetric parameters obtained from the simulations with PRIMO agreed within 2% with the experimental data from IROC-H, except the SBRT-style output factors, which agreed within 3%. Although a fine- tuning is possible with PRIMO to closely match simulation results with a particular linac, the results obtained with the default beam parameters are consistent with the typical values found for 6 MV photon beams from Varian Clinac 2100 linacs.