ESTRO 38 Abstract book

S1185 ESTRO 38

Also, heart D20 was examined as well as D5 for the healthy lung Results Comparison of both calculation methods presented differences in dose distribution in treatment plans. Were noted differences in dose points in distances from applicator axis as well as differences in dose distribution for PTV and OARs. Difference in value of median for the dose point in 1 cm distance from the applicator is 0,2%, maximal value was 41,78% and minimal value was 0,04%. For PVT V 100 , the difference was 2,6% and for the lung D5 was 3,6%. Conclusion Differences in dose distribution in HDR brachytherapy for lung cancer between TG-43 and TG-186 prove that the we should consider the use of algorithms taking into account the tissue mass density. EP-2143 TRAK per unit reference dose as a QA tool is insensitive to finding cervix brachy planning errors P. Gonzalez 1 , F. Koetsveld 1 , A. Mans 1 1 Netherlands Cancer Institute, Radiotherapy, Amsterdam, The Netherlands Purpose or Objective To ensure the quality of a clinical treatment plan for HDR brachytherapy of the cervix, several checks can be made. Following the recommendations of the ICRU report 38, one of these checks is the calculation and reporting of the total reference air kerma (TRAK). This physical quantity, which is defined as the integral of the reference air kerma rate at 1 m distance from the source over the duration of treatment, is believed to be a reliable patient-specific QA tool in determining the treatment intensity. In our clinic the TRAK per unit reference dose, TRAK/D ref , as calculated from the TPS is compared to a predicted value. This predicted value is taken from a linear model relating the treatment volume (TV) to the TPS calculated TRAK/D ref . The TRAK check is passed if the predicted value does not differ more than 3% from the TPS calculated value. The purpose of this work is to test our hypothesis that the TRAK check is not sensitive enough in finding cervix brachy planning errors. Material and Methods The linear model was determined by fitting TRAK/D ref to the treatment volume using a total of 51 clinical intra- cavitary/ interstitial plans. To determine the sensitivity of the TRAK check, several planning errors were introduced into a clinical plan using the Utrecht interstitial CT/MR applicator. All plans were made with the Oncentra treatment planning system. Planning errors were taken from an ASTRO white paper (Pract. Rad. Onc. (2014) 4, 65– 70) and were complemented by several other error sources we check for in our institute. The errors included: wrongful addition of a dwell position, wrong dose optimization and wrong reference dose. Results Table 1 shows the various types of errors introduced and the deviation of the predicted value of the TRAK per unit reference dose from the calculated value. Only 2 out of 13 planning errors were found using the TRAK check: using a wrong reference dose and wrongful positioning of a dwell position. The deviations in these cases were -15.5% and 4.7% respectively. All other deviations fall within 3% of the clinical no-error plan. In the first case it must be noted that this error will go undetected if the wrong reference dose is entered in both the planning and TRAK check, which is likely to occur in a clinical workflow. The second detected error was due to a misplaced dwell position. It was placed far outside of the high-dose region such that it contributed to the TRAK but not to the treatment volume. This planning error was only detected because we used a strict limit of 3%. We also placed additional dwell positions in the high-dose region. Evidently the treatment volume changed but the ratio TRAK/TV remained constant.

Conclusion We conclude that the TRAK check is not a sensitive QA tool since only two very specific planning errors were detected, one of which had a value just above the threshold, and the other could go undetected depending on the workflow. EP-2144 Feasibility of using Micro Silica Bead TLDs for 3D dosimetry in brachytherapy S. Babaloui 1 , S. Jafari 2 , A. L.Palmer 2 , W. Polak 2 , M. W.J.Hubbard 3 , T. Skopidou 3 , A. Lohstroh 3 , R. Jaberi 4 1 Cancer Institute Tehran Univ. of Medical Science, Department of Medical Physics and Biomedical Engineering, Tehran, Iran Islamic Republic of; 2 Portsmouth Hospitals NHS Trust, Medical Physics Dept., Portsmouth, United Kingdom; 3 University of Surrey, Department of Physics, Guildford, United Kingdom ; 4 Cancer Institute- Tehran University of Medical Sciences, Radiation Oncology Research Centre RORC, Tehran, Iran Islamic Republic of Purpose or Objective Evaluation of the feasibility of using silica bead TLDs for 3D phantom dosimetry in brachytherapy (BT) for verification of treatment doses and routine quality control. Estimation of beam perturbation effect of TLDs and clinically relevant spatial resolution (SR) required for placing the beads at the organs’ surface. Material and Methods The TLDs were positioned with 2.5, 5 and 7.5mm spacing on the surface of a 5cc syringe (Fig1a) in a water tank of similar size to the pelvis of a medium patient. CT scans are taken and a 4field conformal radiotherapy (CRT) plan was created using 6MV beams. From the treatment planning system (TPS), 3dose planes were extracted at different depths, in the middle, bottom and below the syringe for every TLD’s arrangement. By using a 2D- γ analysis, the TPS reports were compared with and without presence of the beads. A novel anthropomorphic phantom was developed to position the TLDs. It consists of all pelvic organs which are relevant in gynaecology (GYN) BT. Organs were 3D printed using the CT scan datasets of handmade organs (Fig1 b & d). Acrylonitrile butadiene styrene (ABS) with 90% infill density was used with a CubePro printer (3D Systems Inc., U.S). Bones were made of PVC coated with clinical plaster. The phantom components were confirmed to have mass density and CT numbers similar to the relevant