- Email: [email protected]
28. Optimization of treatment planning parameters used in helical tomotherapy for breast cancer patients and influence on delivery quality assurance
C. Adrien et al. / Physica Medica 44 (2017) 28–45
27. Metallic hip implants: Are avoidance sectors necessary for pelvic VMAT treatments? N. Koutsouvelis, G. Dipasquale, M. Rouzaud, R. Miralbell, T. Zilli Hopitaux Universitaires de Genève (HUG), Rue Gabrielle Perret-Gentil, CH – 1211 Geneva 14, Switzerland Introduction. Avoidance of metallic hip implants (MHI), especially when bilateral, deteriorates the quality of RT treatments, with a significant increase in the dose delivered to the healthy tissues. The aim of this study was to investigate the dosimetric impact of MHI with or without use of avoidance sectors (AS) in classical VMAT planning for pelvic tumors and to assess the in vivo dose calculation errors in the presence of MHI. Methods. In this study we evaluated: (1) The dosimetric impact of AS as assessed on CT planning datasets of four male patients with bilateral MHI (each patient planned for a whole pelvis VMAT to 64 Gy for bladder cancer treatment and for a prostate only VMAT treatment to 78 Gy). No-AS plans: 2 coplanar arcs/ optimization realized as in a classical pelvic VMAT treatment without MHI. AS plans: AS were chosen in a way that no part of the beam passed through the MHI to hit the target. (2) The dose calculation error of a static open beam output through a MHI, using a titan and a ceramic hip prosthesis model in a homemade phantom, 4 different detectors, and 3 different beam energies. (3) The dose calculation error in the PTV during a non-AS 360° VMAT treatment, given the static beam error introduced by the MHI. (4) The dosimetric influence of MHI shifts generated by patient’s repositioning tilts for IGRT purposes. Results. (1) Compared to VMAT plans generated without AS, for all treatment plans, the use of AS resulted in a median increase of the rectal V50Gy and V60Gy by 424% and 339% respectively, and deteriorated the conformity. For prostate treatment plans, a median increase of 56%, 48% and 19% was observed for bladder V60Gy, V65Gy, and V75Gy respectively, when using AS. (2) For distances between 0.5 cm and 6 cm after the MHI, the deviations of measurements from calculations were not significant.
39
On the surface of the MHI, significant deviations 12% and 14% were observed with EBT3 films, for the CHP and the titanium material respectively (only X6 beam tested). (3) For a given static open beam error (point2), introduced by the MHI in the PTVs, the respective 360° VMAT treatments showed a decrease of this error by a median factor of 4 (Annexe). (4) Patient’s tilt must not lead in a MHI anterior-posterior shift of more than 4 mm, in order to have a maximum error of 1% at the PTV borders. Conclusions. In patients with bilateral MHI treated with pelvic 360° VMAT, creation of AS for treatment planning is not mandatory, especially when the distance between the prosthetic material and the target volume is more than 5 mm. Planning improvement in terms of simplicity, conformality and organ sparing is expected by irradiating through MHI when using VMAT techniques. https://doi.org/10.1016/j.ejmp.2017.10.107
28. Optimization of treatment planning parameters used in helical tomotherapy for breast cancer patients and influence on delivery quality assurance C. Adrien, L. Bartolucci, A. Mazal Institut Curie, Paris, France Introduction. Helical tomotherapy treatment planning is a multistep process where unique parameters need to be set: pitch (p), field width (FW) and modulation factor (MF). Depending on these parameters both plan quality and execution time can be affected. At our institution, treatment planning for breast cancer with lymph node involvement is standardized and meets stringent organs at risk (OAR) dose constraints. The aim of this study was to optimize these treatment plans by changing the following parameters used in routine clinical practice: FW = 2.5 cm, MF = 2.5 and p = 0.287. Methods. 10 patients treated for breast cancer with lymph node involvement were selected. Clinically approved plans were first reproduced with a fixed number of iteration in order to compare
40
J. Akunzi et al. / Physica Medica 44 (2017) 28–45
plans among themselves on the same level (structures and constrains remaining untouched). Then, two plan parameters were changed: MF (2.5 and 3) and pitch based on Kissick et al. and Chen et al. published data (0.287, 0.420, 0.436) [1,2]. Artiview software (Aquilab) was used to evaluate and compare plans using dosimetric indices for OAR and target volumes. Specific patient delivery quality assurance (DQA) were finally performed using the MatriXX detector array (IBA). DQA for all patients were carried out in the same condition as the original one approved by the Medical Physics team. Paired Student’s t-test (a = 0.05) was used to cross-check all plans. Results. Results show that increasing MF from 2.5 to 3 improves target volumes coverage and OAR sparing. Yet, these improvements are linked with prolonged treatment time and a degradation of DQA passing rates. Adapted pitches according to off-axis distance [2] seem to provide equivalent plans as thus provided with our standard pitch [1], with the added benefit of reducing stressed on the multileaf collimator. Conclusions. First results are promising and should lead us in improving our clinical practice. Further work will involve the study of the 5 cm field width collimation. References
dose detector, a water phantom and a small solid water phantom to hold this detector, a point dose measurement is insufficient to assess the geometric precision of the dose-fall off during arc delivery. Given the safety requirements for stereotactic treatments, it is therefore highly recommended to invest in a detector system that can provide 2D and 3D dose information as well. The 1000SRS was found to provide very reliable planar dose measurements and, in combination with the Octavius4D system, measurement-based 3D dose reconstructions. It is also the most efficient method, especially when multiple lesions are concerned. From the battery of validation measurements, it was found that, although the algorithm configuration as well as the MLC modeling within the Eclipse TPS could benefit from further improvements, the currently obtained results are within clinical acceptance for the specific requirements of stereotactic treatment plans. Conclusions. Target localization remains the key aspect of successful stereotactic radiotherapy and should be carefully addressed according to the treatment site. However, from a dosimetric point of view, when the appropriate measurement equipment is available, safe implementation of stereotactic RA treatments should be within reach of all radiotherapy departments outfitted with an up to date Clinac (or TrueBeam) and state-of-the-art on-board imaging.
1. Kissick et al.. Med Phys(32). 2. Chen et al.. Med Phys(38).
https://doi.org/10.1016/j.ejmp.2017.10.109
https://doi.org/10.1016/j.ejmp.2017.10.108
30. Monte Carlo modeling Varian TrueBeam Novalis STX linear accelerator and methods of validation J. Akunzi a, P.E. Leni b, R. Gschwind c
29. Implementing stereotactic RapidArc treatments into clinical routine: From algorithm configuration to treatment validation F. Sergent a, A. Van Esch b, K. Basta c, P. Bertrand d, C. Corbice d, E. Fontaine d, L. Hambach a, C. Clermont a, A. Blavier a, D. Huyskens b a
CHU UCL Namur, Place Louise Godin, 15 5000 Namur, Belgium 7sigma, Kasteeldreef 2, 3150 Tildonk, Belgium c Centre hospitalier de Mouscron, Avenue de Fécamp 49, 7700 Mouscron, Belgium d CHU de La Réunion, Av. Francois Mitterand BP305, 97448 Saint-Pierre, Reunion b
Introduction. This work aims to aid the medical physicist with the safe implementation of RapidArc (RA) (Varian Medical Systems, Palo Alto, CA) stereotactic radiotherapy treatments (SRS/SBRT) into clinical routine, from treatment planning system (TPS) configuration to patient plan verification. Implementation procedures are applicable to different Varian linear accelerators, either equipped with a standard Millennium 120MLC or a high-definition HDMLC, but always with on-board imaging. Methods. A systematic approach was used to assure proper control of the different aspects of the implementation. First, an extensive series of detectors (all from PTW, Freiburg, Germany) – from numerous point dose detectors to the 1000SRS/Octavius4D 3D dose measurement system - were carefully benchmarked to assess their dosimetric characteristics, their precision and their practical usefulness. This benchmarking was performed independently of the TPS. Second, the necessary measurements were performed to include small field data in the Analytical Anisotropoic Algorithm (AAA) and Acuros (AXB) algorithm configuration. Third, validation of the Eclipse small field dose calculation was performed for both algorithms, starting off with static gantry (small) MLC fields and ending with RA SRS/SBRT test plans. Finally, pre-treatment QA procedures were implemented, executed and analyzed on all patient treatments. Results. While one can do a substantial part of the basic validation with a single, high resolution, directionally independent point
a
Centre Antoine Lacassagne, Nice – CRLCC Antoine Lacassagne – 33, avenue de Valombrose, 06189 Nice cedex 2, France b IEI-FC, Université de Franche-Comté, Technopole TEMIS, 18 rue Alain Savary 25000 Besançon, France c IRMA/CE, CNRS UMR 6249, Université de Franche-Comté, Montbéliard (UMR 6249 – Laboratoire Chrono-environnement), Université de Franche-Comté – Pôle Universitaire BP 71427, 25211 Montbéliard cedex, France Introduction. In order to safely administrated highly complexes radiotherapy technics like VMAT or IMRT, the implementation of a quality assurance software is in progress at Chrono-environnement UMR CNRS 6249 laboratory based in Franche-Comté (France). The Monte Carlo modeling of the TrueBeam Novalis STx used in Montbéliard hospital was performed. Hereafter, the validation for clinic use of IAEA-compliant phase-space files for 10X, and 10XFFF. These phase-space files were calculated by the Varian Monte Carlo research team. Methods. The modeling of TrueBeam Novalis STX of Varian was implemented using the Monte Carlo code BEAMnrc/DOSXYZnrc. In order to validate the modeling, the simulations of the phase-space files for the 10 MV beams in both flattened and unflattened (FFF) mode were performed. BEAMnrc was used to create field specific phase spaces under the jaws for field ranging from 3 cm ⁄ 3 cm to 20 cm ⁄ 20 cm. DOSXYZnrc was used to calculate doses in water phantom. Calculated cross-lines profiles, percent depth doses (PDD) and output factors were compared with measurements provided by Montbéliard hospital. The c-index test was performed on both PDDs and cross-lines profiles. Results. For cross-line profiles, agreements in the 80%–20% width and in the 50% field size were better than 2 mm for all fields. More over agreement better than 1.5%, 1 mm and 2%, 2 mm were found for 10 MV FFF and 10 MV, respectively. Calculated and measured percent depth doses beyond the buildup region agreed within 1:5%, 1 mm for 10 MV FFF beams and within 2%, 2 mm for 10 MV
27. Metallic hip implants: Are avoidance sectors necessary for pelvic VMAT treatments? N. Koutsouvelis, G. Dipasquale, M. Rouzaud, R. Miralbell, T. Zilli Hopitaux Universitaires de Genève (HUG), Rue Gabrielle Perret-Gentil, CH – 1211 Geneva 14, Switzerland Introduction. Avoidance of metallic hip implants (MHI), especially when bilateral, deteriorates the quality of RT treatments, with a significant increase in the dose delivered to the healthy tissues. The aim of this study was to investigate the dosimetric impact of MHI with or without use of avoidance sectors (AS) in classical VMAT planning for pelvic tumors and to assess the in vivo dose calculation errors in the presence of MHI. Methods. In this study we evaluated: (1) The dosimetric impact of AS as assessed on CT planning datasets of four male patients with bilateral MHI (each patient planned for a whole pelvis VMAT to 64 Gy for bladder cancer treatment and for a prostate only VMAT treatment to 78 Gy). No-AS plans: 2 coplanar arcs/ optimization realized as in a classical pelvic VMAT treatment without MHI. AS plans: AS were chosen in a way that no part of the beam passed through the MHI to hit the target. (2) The dose calculation error of a static open beam output through a MHI, using a titan and a ceramic hip prosthesis model in a homemade phantom, 4 different detectors, and 3 different beam energies. (3) The dose calculation error in the PTV during a non-AS 360° VMAT treatment, given the static beam error introduced by the MHI. (4) The dosimetric influence of MHI shifts generated by patient’s repositioning tilts for IGRT purposes. Results. (1) Compared to VMAT plans generated without AS, for all treatment plans, the use of AS resulted in a median increase of the rectal V50Gy and V60Gy by 424% and 339% respectively, and deteriorated the conformity. For prostate treatment plans, a median increase of 56%, 48% and 19% was observed for bladder V60Gy, V65Gy, and V75Gy respectively, when using AS. (2) For distances between 0.5 cm and 6 cm after the MHI, the deviations of measurements from calculations were not significant.
39
On the surface of the MHI, significant deviations 12% and 14% were observed with EBT3 films, for the CHP and the titanium material respectively (only X6 beam tested). (3) For a given static open beam error (point2), introduced by the MHI in the PTVs, the respective 360° VMAT treatments showed a decrease of this error by a median factor of 4 (Annexe). (4) Patient’s tilt must not lead in a MHI anterior-posterior shift of more than 4 mm, in order to have a maximum error of 1% at the PTV borders. Conclusions. In patients with bilateral MHI treated with pelvic 360° VMAT, creation of AS for treatment planning is not mandatory, especially when the distance between the prosthetic material and the target volume is more than 5 mm. Planning improvement in terms of simplicity, conformality and organ sparing is expected by irradiating through MHI when using VMAT techniques. https://doi.org/10.1016/j.ejmp.2017.10.107
28. Optimization of treatment planning parameters used in helical tomotherapy for breast cancer patients and influence on delivery quality assurance C. Adrien, L. Bartolucci, A. Mazal Institut Curie, Paris, France Introduction. Helical tomotherapy treatment planning is a multistep process where unique parameters need to be set: pitch (p), field width (FW) and modulation factor (MF). Depending on these parameters both plan quality and execution time can be affected. At our institution, treatment planning for breast cancer with lymph node involvement is standardized and meets stringent organs at risk (OAR) dose constraints. The aim of this study was to optimize these treatment plans by changing the following parameters used in routine clinical practice: FW = 2.5 cm, MF = 2.5 and p = 0.287. Methods. 10 patients treated for breast cancer with lymph node involvement were selected. Clinically approved plans were first reproduced with a fixed number of iteration in order to compare
40
J. Akunzi et al. / Physica Medica 44 (2017) 28–45
plans among themselves on the same level (structures and constrains remaining untouched). Then, two plan parameters were changed: MF (2.5 and 3) and pitch based on Kissick et al. and Chen et al. published data (0.287, 0.420, 0.436) [1,2]. Artiview software (Aquilab) was used to evaluate and compare plans using dosimetric indices for OAR and target volumes. Specific patient delivery quality assurance (DQA) were finally performed using the MatriXX detector array (IBA). DQA for all patients were carried out in the same condition as the original one approved by the Medical Physics team. Paired Student’s t-test (a = 0.05) was used to cross-check all plans. Results. Results show that increasing MF from 2.5 to 3 improves target volumes coverage and OAR sparing. Yet, these improvements are linked with prolonged treatment time and a degradation of DQA passing rates. Adapted pitches according to off-axis distance [2] seem to provide equivalent plans as thus provided with our standard pitch [1], with the added benefit of reducing stressed on the multileaf collimator. Conclusions. First results are promising and should lead us in improving our clinical practice. Further work will involve the study of the 5 cm field width collimation. References
dose detector, a water phantom and a small solid water phantom to hold this detector, a point dose measurement is insufficient to assess the geometric precision of the dose-fall off during arc delivery. Given the safety requirements for stereotactic treatments, it is therefore highly recommended to invest in a detector system that can provide 2D and 3D dose information as well. The 1000SRS was found to provide very reliable planar dose measurements and, in combination with the Octavius4D system, measurement-based 3D dose reconstructions. It is also the most efficient method, especially when multiple lesions are concerned. From the battery of validation measurements, it was found that, although the algorithm configuration as well as the MLC modeling within the Eclipse TPS could benefit from further improvements, the currently obtained results are within clinical acceptance for the specific requirements of stereotactic treatment plans. Conclusions. Target localization remains the key aspect of successful stereotactic radiotherapy and should be carefully addressed according to the treatment site. However, from a dosimetric point of view, when the appropriate measurement equipment is available, safe implementation of stereotactic RA treatments should be within reach of all radiotherapy departments outfitted with an up to date Clinac (or TrueBeam) and state-of-the-art on-board imaging.
1. Kissick et al.. Med Phys(32). 2. Chen et al.. Med Phys(38).
https://doi.org/10.1016/j.ejmp.2017.10.109
https://doi.org/10.1016/j.ejmp.2017.10.108
30. Monte Carlo modeling Varian TrueBeam Novalis STX linear accelerator and methods of validation J. Akunzi a, P.E. Leni b, R. Gschwind c
29. Implementing stereotactic RapidArc treatments into clinical routine: From algorithm configuration to treatment validation F. Sergent a, A. Van Esch b, K. Basta c, P. Bertrand d, C. Corbice d, E. Fontaine d, L. Hambach a, C. Clermont a, A. Blavier a, D. Huyskens b a
CHU UCL Namur, Place Louise Godin, 15 5000 Namur, Belgium 7sigma, Kasteeldreef 2, 3150 Tildonk, Belgium c Centre hospitalier de Mouscron, Avenue de Fécamp 49, 7700 Mouscron, Belgium d CHU de La Réunion, Av. Francois Mitterand BP305, 97448 Saint-Pierre, Reunion b
Introduction. This work aims to aid the medical physicist with the safe implementation of RapidArc (RA) (Varian Medical Systems, Palo Alto, CA) stereotactic radiotherapy treatments (SRS/SBRT) into clinical routine, from treatment planning system (TPS) configuration to patient plan verification. Implementation procedures are applicable to different Varian linear accelerators, either equipped with a standard Millennium 120MLC or a high-definition HDMLC, but always with on-board imaging. Methods. A systematic approach was used to assure proper control of the different aspects of the implementation. First, an extensive series of detectors (all from PTW, Freiburg, Germany) – from numerous point dose detectors to the 1000SRS/Octavius4D 3D dose measurement system - were carefully benchmarked to assess their dosimetric characteristics, their precision and their practical usefulness. This benchmarking was performed independently of the TPS. Second, the necessary measurements were performed to include small field data in the Analytical Anisotropoic Algorithm (AAA) and Acuros (AXB) algorithm configuration. Third, validation of the Eclipse small field dose calculation was performed for both algorithms, starting off with static gantry (small) MLC fields and ending with RA SRS/SBRT test plans. Finally, pre-treatment QA procedures were implemented, executed and analyzed on all patient treatments. Results. While one can do a substantial part of the basic validation with a single, high resolution, directionally independent point
a
Centre Antoine Lacassagne, Nice – CRLCC Antoine Lacassagne – 33, avenue de Valombrose, 06189 Nice cedex 2, France b IEI-FC, Université de Franche-Comté, Technopole TEMIS, 18 rue Alain Savary 25000 Besançon, France c IRMA/CE, CNRS UMR 6249, Université de Franche-Comté, Montbéliard (UMR 6249 – Laboratoire Chrono-environnement), Université de Franche-Comté – Pôle Universitaire BP 71427, 25211 Montbéliard cedex, France Introduction. In order to safely administrated highly complexes radiotherapy technics like VMAT or IMRT, the implementation of a quality assurance software is in progress at Chrono-environnement UMR CNRS 6249 laboratory based in Franche-Comté (France). The Monte Carlo modeling of the TrueBeam Novalis STx used in Montbéliard hospital was performed. Hereafter, the validation for clinic use of IAEA-compliant phase-space files for 10X, and 10XFFF. These phase-space files were calculated by the Varian Monte Carlo research team. Methods. The modeling of TrueBeam Novalis STX of Varian was implemented using the Monte Carlo code BEAMnrc/DOSXYZnrc. In order to validate the modeling, the simulations of the phase-space files for the 10 MV beams in both flattened and unflattened (FFF) mode were performed. BEAMnrc was used to create field specific phase spaces under the jaws for field ranging from 3 cm ⁄ 3 cm to 20 cm ⁄ 20 cm. DOSXYZnrc was used to calculate doses in water phantom. Calculated cross-lines profiles, percent depth doses (PDD) and output factors were compared with measurements provided by Montbéliard hospital. The c-index test was performed on both PDDs and cross-lines profiles. Results. For cross-line profiles, agreements in the 80%–20% width and in the 50% field size were better than 2 mm for all fields. More over agreement better than 1.5%, 1 mm and 2%, 2 mm were found for 10 MV FFF and 10 MV, respectively. Calculated and measured percent depth doses beyond the buildup region agreed within 1:5%, 1 mm for 10 MV FFF beams and within 2%, 2 mm for 10 MV