S Stathakis

SU‐E‐T‐361: Evaluation of a New Commercially Device for Patient Specific IMRT QA

Purpose: To evaluate the performance of a new system for patient specific IMRT QA. Methods: The new Octavius 4D system and its accompanying software (Verisoft) by PTW were used to compare calculated and measured dose distributions for ten (n=10) patients plans. Five IMRT and five SmartArc plans were used for the comparison. All plans were optimized using the Pinnacle treatment planning system and the plans and their respective verification plans were calculated using collapsed cone convolution superposition dose calculation algorithm. The verification plans were evaluated using the gamma index and also by visually inspecting the dose profiles and isodose distributions. The Verisoft IMRT QA is able to reconstruct the 3D dose from individual planar measurements and hence provide comparison for transverse, sagittal, and coronal planes. Results: The gamma index values varied from 92.1 to 99.2% when the coronal plan was only evaluated. Because of the ability of the system to compose thedose distribution from the individual measurements, comparison of the sagittal, transverse planes as well as 3d dose distributions were available. The gamma index values varied from 88.3 to 99.5% for different planes while the 3D gamma index values ranged from 85.0 to 99.5%. Conclusions: The new phantom for patient specific IMRT QA in this study can effectively be used for clinical use. The ability of the software to compute 3D dose distribution from all measurements taken is powerful tool that aids the physicist to evaluate the verification plan by providing 3D information not available from other commercial systems. Further investigation of the systems dose reconstruction algorithm is necessary to determine its accuracy. This work was partially funded with an educational grant from PTW.

SU-E-T-341: Characterization of a New Commercially Available System for Patient Specific IMRT/VMAT QA

Purpose: To evaluate and characterize the 3D dose reconstruction algorithm of a new commercially available system for patient specific IMRT QA. Methods: champalimaud The new Octavius 4D system and its accompanying software (Verisoft) by PTW were evaluated for the dose reconstruction accuracy. OCTAVIUS 4D measures the dose in the isocenter. The phantom rotates synchronously with the gantry by means of an inclinometer and a motor. The collected measurements are then reconstructed to a 3D dose volume for comparison with the calculated dose in the phantom. In this study, we evaluated the 3D dose reconstruction algorithm of the system using a series of tests. Dose distributions for various field sizes, orientations, shapes and combination of fields were calculated using the Pinnacle TPS on the phantom and the respective DICOMRT dose exported to the Verisoft. Example of these fields included 3×3cm2 to 25×25cm2, wedged fields, AP/PA arrangements, as well as lateral fields and IMRT fields. The respective measurements were performed and comparisons were made based on gamma index, profiles and isodose distributions. Results: The 3D gamma index ranged from 88.3 to 99.9%. Individual dose planes (axial, sagittal, and coronal) were also evaluated and their gamma was within the 3D gamma range. The new system independent of angular dependence as it rotates with gantry and the measuring array is at constant SAD and perpendicular to the beams axis. Conclusions: The new phantom for patient specific IMRT QA in this study can effectively be used for clinical use. The ability of the software to compute 3D dose distribution from all measurements taken is powerful tool that aids the physicist to evaluate the verification plan by providing 3D information not available from other commercial systems. This work has been partially funded by an education grant from PTW.

WE‐G‐BRCD‐05: Evaluation of Localization Errors for CSA Delivery Using VMAT

PURPOSEnTo dosimetrically evaluate the effects of improper patient positioning in the junction area of a VMAT cranio-spinal axis irradiation technique consisting of one superior and one inferior arc.nnnMETHODSnFive (n=5) cranio-spinal axis irradiation patients were planned with 2 arcs: one superior and one inferior. The plans were then recalculated with inferior isocenter shifted, in order to mimic patient setup errors, eight times: lmm, 2mm, 5mm, and 10mm superiorly, and 1mm, 2mm, 5mm, and 10mm inferiorly. Plans were then compared to the original, non-shifted arc plan based on target metrics of conformity number and homogeneity index, as well as several normal structure mean doses.nnnRESULTSnPercent differences were calculated in order to compare each of the eight shifted plans to the original arc plan without shifts, which would be the ideal setup of patient without error. The conformity number was on average 0.87%, 2.74%, 5.75%, and 9.10% lower for the 1mm, 2mm, 5mm, and 10mm inferiorly- shifted plans and 0.41%, 0.82%, 2.75%, and 5.99% lower for the respective superiorly-shifted plans. The homogeneity indices were, averaged among the five patients, 0.03%, 0.26%, 0.97%, and 2.84% for the inferior shifts and 0.23%, 1.17%, 6.31%, and 15.29% worse, or less homogenous for the superior shifts. Overall the mean doses to the organs at risk were less than 2% different for the 1mm, 2mm, and 5mm shifted plans. The 10mm shifted plans, however, showed percent differences from original plan of up to 5.6% on average.nnnCONCLUSIONSnSetup errors when shifting isocenters should be minimized in order to provide the patient with the best treatment possible. Errors of 1 to 2mm can negatively affect patient treatment, most notably in the arc junction area, but are not as problematic as larger errors of 5 to 10mm.

SU‐E‐T‐562: Reducing the Arc Span for CSA VMAT Delivery

PURPOSEnTo dosimetrically evaluate advantages and disadvantages of using multiple, shorter, sub-arcs versus full arc deliveries for treatment of cranio-spinal axis (CSA) irradiation.nnnMETHODSnFive (n=5) cranio-spinal axis irradiation patients were planned using 2 complete arcs, one superior and one inferior; with gantry rotations from 1 to 359 degrees. Due to supine patient setup, each original full arc was then replanned split into two sub arcs with gantry rotations from 1 to 100 and 260 to 359 degrees creating 4 smaller arcs. The PTV was normalized such that 95% received at least 23.4 Gy in 13 fractions. The PTV was evaluated based on conformity number and homogeneity index. The normal structures were evaluated based on maximum and mean doses. Beam on times and monitor units were compared.nnnRESULTSnAveraged over all patients, conformity number was calculated to be approximately 0.86 and 0.82 for full arc and sub arc plans respectively. The homogeneity index was approximately 1.07 and 1.06 for full arc and sub arc plans. This indicates better target conformity but less homogeneous dose distribution for full arc plans as compared with sub arc plans. With the exception of the eyes, each normal structure evaluated had lower maximum doses with subarc plans. All normal structures, with the exception of the left kidney, had lower mean doses using sub arc deliveries. Beam on times were shorter on average for full arcs, but the monitor units were lower on average for sub arcs.nnnCONCLUSIONSnOverall, CSA patients would benefit from the use of sub arc treatment deliveries versus full arc deliveries. Nearly all normal structure doses were lower for sub arcs, while the PTV was still adequately covered and beam on times and monitor units were similar.

SU‐E‐T‐465: Exploring the Dosimetric and Tumor Control Consequences of Prostate Seed Loss and Migration

PURPOSEnOnce implanted, prostate brachytherapy seeds are vulnerable to loss and movement. A general estimation of these effects may be useful for making patient care decisions when seeds are lost after the post-implant scan. The goal of this work was to explore the dosimetric and radiobiological effects of the types of seed loss and migration common in prostate brachytherapy.nnnMETHODSnThis study evaluates five patients. For each, three treatment plans were created using Iodine-125, Palladium-103 and Cesium-131. The three seeds closest to the urethra were identified and modeled as seeds lost through the urethra. The three seeds closest to the exterior of prostatic capsule were identified and modeled as those lost from the prostate periphery. The seed locations and organ contours were exported from Prowess and used by in-house software to perform the dosimetric and radiobiological evaluation. The radiobiological evaluation was based on the linear-quadratic model. Seed loss was simulated by removing 1, 2 or 3 seeds near the urethra 0, 2 or 4 days after the implant or removing seeds near the exterior of the prostate 14, 21 or 28 days after the implant.nnnRESULTSnLoss of 1, 2 or 3 seeds through the urethra resulted in D90 reduction of 2%, 5% and 7% loss respectively. Due to delayed loss of peripheral seeds, effects were less severe than for loss through the urethra. However, while the dose reduction is modest for multiple lost seeds, the reduction in tumor control probability was minimal.nnnCONCLUSIONSnThe goal of this work was to explore the dosimetric and radiobiological effects of the types of seed loss and migration commonly seen in prostate brachytherapy. The results presented show that loss of multiple seeds can cause a substantial reduction of D90 coverage. However, the dose reduction was not seen to significantly reduce tumor control probability.

SU‐E‐T‐91: VMAT Monthly Quality Assurance Using an in Vivo Dosimetric Device Attached to the Linac Gantry

Purpose: Varian RapidArc is a volumetric arc therapy (VMAT) that obtains a conformal dose around the desired structure by employing variable gantry speed, dose rate, and dynamic multileaf collimator (DMLC) speed as the gantry rotates about machine isocenter. This study is meant to build upon previous research by Ling et. al. by completing the tests with an in vivo dosimetric device attached to the linac gantry. Methods: The PTW DAVID detector was attached to the linear accelerator gantry, which allows it to remain perpendicular to the beam at all gantry angles. Three tests for RapidArc evaluation were performed on this device including: dose rate and gantry speed variation, DMLC speed and dose rate variation, and DMLC position accuracy. The reproducibility of the arc data was also reported. Results: A picket fence plan varying dose rates from 111 to 600 MU/min and gantry speeds from 5.5 to 4.3 degrees/s was delivered consisting of seven sections of different combinations. These measurements were compared to static gantry, open field measurements and found to be within 2.39% (2.23–2.39%). A four‐section picket fence for DMLC speeds of 0.46, 0.92, 1.84, and 2.76 cm/s was similarly evaluated and found to be within 1.99% (1.86–1.99%). For DMLC position accuracy s picket fence arc plan was compared to a static picket fence and found to agree within 0.38% (0.31–0.38%). Reproducibility for these three RapidArc plans was found to be within 0.19%, 0.16%, and 0.30% respectively. Conclusions: The DAVID detector was able to measure the RapidArc quality assurance plans accurately and was found to produce reproducible data. Testing the main three elements of variation for RapidArc delivery is a necessary component of VMAT evaluation and this device allows for a time efficient method of doing so. This work has been partially funded by PTW, Freiburg, Germany.

SU‐E‐T‐702: Accuracy of a Commercially Available Dose Calculation Algorithm for Small Field Dosimetry

Purpose: To validate the AcurosXB dose calculation algorithm for small field dosimetry for high energy photon beams. Methods: Calculations in water phantom of 0.2×0.2×0,2cm3 voxel size with slabs of inhomogeneity materials (air rho=0.0012g/cm3, lung rho =0.2 g/cm3 and bone rho =1.85 g/cm3) were used in this study. Fields sizes ranging from 1×1cm2 to 10×10cm2 and energies of 6, 10 and 18MV were calculated using the Acuros XB dose calculation algorithm available in the Eclipse treatment planning system (TPS). The calculations were compared to the Analytical Anisotropic Algorithm (AAA) also available in Eclipse as well to Collapsed Cone Convolution Superposition (CCCS) algorithm available in the Pinnacle3 TPS. Comparison between the dose calculations and Monte Carlo calculations using EGSnrcBEAMnrc and EGSnrcDOSXYZnrc package was also performed. Results: The AcurosXB calculation algorithm was in general in good agreement with Monte Carlo calculations. Discrepancies were observed at the interfaces of the inhomogeneities. Good agreement between CCCS and AcurosXB was also observed for the majority of the cases with discrepancies observed only at the interfaces of the media. The AAA did not successfully calculate the dose for the test geometries when small fields were used. Discrepancies in the order of 50–70% were observed. The AAA overestimated the dose in the low density material and failed to predict the second buildup region accurately for the very small field sizes. Conclusion: In general, the overall degree of accuracy for AcurosXB in the conditions of electronic disequilibrium was in good agreement with Monte Carlo calculations (within 2%) and comparable with the CCCS algorithm. The AAA on the other hand failed to accurately predict the dose for the small fields studied in the presence of inhomogeneities

SU‐E‐T‐549: Hypofractionated Spine Radiotherapy: How Do You QA?

Purpose: To evaluate the ability of commercially available devices to perform IMRT QA measurements for plans involving small volume targets as in the case of single spine vertebra radiotherapy Methods: Seven (n=7) patient with metastatic spine lesions were treated using RapidArc plans. The plans were optimized so that a single arc was to deliver the dose of 4.0Gy per fraction for 5 fractions to the vertebral body while limiting the dose to the spinal cord. For each approved plan four (n=4) verification plans were developed corresponding to four different IMRT QA detectors (EBT2 film, IBA MatrixX, PTW Seven29, and Scandidos Delta4) employed for the evaluation of the patient IMRT QA. Each QA was evaluated for its agreement against the verification plan using the gamma index, profiles and planar isodose distribution agreement. Results: The Scandidos Delta4 device had the best agreement. The PTW Seven29 detector array had similar response to film measurements. The response of the PTW Seven29 array improved when the multiple, off centered measurements were averaged. The IBA MatriXX device had inferior response compared to the other detector arrays and had the lowest gamma passing ratesConclusion: The choice of the IMRT QA device has an important role in the results of the patient specific quality assurance of the delivered dose to the patient in the case of small targets as in the treatment of spinal targets.

SU‐E‐T‐36: Sensitivity of Tumor Control Probability to Changes in Radiobiological Parameters

Purpose: Radiobiological parameters used to calculate tumor control probabilities (TCP) are known to vary among patients. It then becomes important to explore the effects different parameters have on TCP calculations. Here, the effect of different parameter combinations on TCP for several prostate seed implant treatments was explored. Methods: Forty‐four patients were evaluated. All patients were treated with Pd‐103 seeds and were planned on the Spot system. The prescription dose was 120 Gy. A post‐implant CT was taken for each patient and was used to create a tumor DVH. Using this information and different biological parameters, the linear‐quadratic model was used to calculate the biological effective dose (BED). The BED was then used to calculate the TCP. Combinations of the following biological parameters were evaluated: alpha/beta = 2, 3, 4 Gy; D50 = 30, 50, 70 Gy; and maximum normalized dose‐response gradient (gamma) = 2, 3, 4. The TCP for each patient was calculated for each of the twenty‐seven different combinations.Results: Combinations with the highest TCP at a given low BED were those were alpha/beta and gamma were high and D50 was low. Combinations with the lowest TCP at a given high BED levels were those were alpha/beta and gamma were low and D50 was high. Parameter values of alpha/beta = 2 Gy, D50 = 50 Gy and gamma = 4 provided TCP values that are consistent with realistic patient outcomes. These parameters yield an α value that was consistent with newer findings Conclusions: In this work, the sensitivity of TCP calculations on choice of biological parameters was evaluated for prostate seed implant patients. This was done by calculating the BED and TCP for the tumor. The parameters that best matched realistic clinical results were alpha/beta = 2 Gy, D50 = 50 Gy and gamma = 4.

SU‐E‐T‐125: VMAT Monthly Quality Assurance Using a 2D Ionization Chamber Array

Purpose: Periodic quality assurance is essential for the accurate and consistent operation of dynamic multileaf collimators (DMLC) for volume modulated arc therapy (VMAT). Varian RapidArc is a system for delivering VMAT that can vary gantry speed and dose rate as the DMLCs quickly change position in order to obtain a conformal dose. A more time efficient and reproducible method for verifying the RapidArc capabilities has been developed building upon previous work by Ling et. al. using a 2D ionization chamber array. Methods: The PTW seven29 2D‐ARRAY was used in conjunction with an isocentric gantry mount, removing directional dependence, to complete three tests. Tests included evaluating machine performance while varying dose rate and gantry speed, while varying DMLC speed and the dose rate, and DMLC position accuracy. Reproducibility of each test was also evaluated. Results: Dose rate (111–600 MU/min) and gantry speed (5.5–4.3 degrees/s) capability was studied with a RapidArc picket fence pattern consisting of 7 different sections of combinations and compared to a corresponding open field and was found to agree within 0.84%. DMLC speed (0.46, 0.92, 1.84, and 2.76 cm/s) evaluation was performed similarly and agreed within 3.66% (2.93–3.66%). A RapidArc picket fence was compared with a static gantry picket fence to evaluate DMLC position accuracy to within 2.91%. Reproducibility of each of the three RapidArc tests was found to be within 0.57%, 1.78%, and 2.70% respectively. Conclusions: The seven29 detector was able to perform the RapidArc quality assurance tests efficiently and accurately and the results were reproducible. Periodic verification of DMLC movement, dose rate variation, and gantry speed variation relating to RapidArc delivery can be completed in a timelier manner using this equipment. This work partially supported by PTW, Freiburg, Germany.