- Email: [email protected]
72 oral A treatment planning comparison of 3D conformal therapy, intensity modulated photon therapy and proton therapy for treatment of advanced head and neck tumours
$32
72
Tuesday, 18 September 2001
oral
A treatment planning comparison of 3D conformal therapy, intensity modulated photon therapy and proton therapy for treatment of advanced head and neck tumours L. Cozzi, A. Fogliata, A. Bolsi IOSI, Med. Phys., Bellinzona, CH
Background and Purpose: In this work, the potential benefits and limitations of different treatment techniques, based on mixed photon-electron beams, 3D conformal therapy, intensity modulated photons (IM) and protons (passively scattered and spot scanned), have been assessed using comparative treatment planning methods in a cohort of patients presenting with advanced head and neck tumours. Material and Methods: Plans for five patients were computed for all modalities using CT scans to delineate target volume (PTV) and organs at risk (OAR) and to predict dose distributions. The prescribed dose to the PTV was 54 Gy, whilst the spinal cord was constrained to a maximum dose of 40.5 Gy for all techniques. Dose volume histograms were used for physical and biological evaluation, which included equivalent uniform dose (EUD) calculations. Results: Excluding the mixed photon-electron technique, PTV coverage was within the defined limits for all techniques, with protons providing significantly improved close homogeneity, resulting in correspondingly higher EUD results. For the spinal cord, protons also provided the best sparing with maximum doses as low as 17 Gy. Whilst the IM plans were demonstrated to be significantly superior to non modulated photon plans, they were found to be inferior to protons for both criteria. A similar result was found for the parotid glands. Although they are partially included in the treated volume there is a clear indication that protons, and to a lesser extent IM photons, could play an important role in preserving organ functionality with a consequent improvement of the patient's quality of life. Conclusions: For advanced head and neck tumours, we have demonstrated that the use of IM photons or protons both have the potential to reduce the possibility of spinal cord toxicity. In addition, a substantial reduction of dose to the parotid glands through the use of protons enhances the interest for such a treatment modality in cases of advanced head and neck tumours. However, in terms of target coverage, the use of 3D conformal therapy, although somewhat inferior in quality to protons or IM photons, has been shown to be a reasonable alternative to the more advanced techniques. In contrast, the conventional technique of mixed photon and electron fields has been shown to be inferior to all other techniques for both target coverage and OAR involvement. 73
oral
Intensity modulated radiotherapy for oropharyngeal cancer: a class solution B. van A~selen, H. Dehnad, C.P.J. Raaijmakers, C.H.J. Terhaard, J.J.W. Lagendijk U.M.C. Utrecht, Department of Radiotherapy, Utrecht, The Netherlands
Purpose: Using intensity modulated radiotherapy (IMRT), it might be possible to increase the dose to the gross tumor volume (GTV) of the oropharynx in order to enhance the tumor control probability (TCP). At the same time the normal tissue complication probability should be lower or similar to that obtained for the conventional technique. The aim of this study is to obtain a class solution, i.e. a standardized treatment planning procedure, for the IMRT treatment of the oropharynx. Methods: Ten patients underwent a planning CT scan. The clinical target volume (CTV) of the lymph nodes, the CTV and GTV of the primary target and the critical organs (myelum, brain and parotid glands) were delineated. A 5 mm margin was added to the CTV to define the planning target volume (PTV). The IMRT plans were generated using an inverse planning system. Sequencing was done in a 'step and shoot' mode using a sequencer developed at our institute. After sequencing, the dose calculations for IMRT plans were performed using a 3D calculation algorithm, which was also used to for the conventional plans. In order to find the class solution, the number of beams, the number of intensity levels and the optimal set of dose constraints (penalty settings) were investigated. Results: Preliminary results show that it is possible to obtain a class solution for the IMRT treatment of the oropharynx tumors. By systematically varying the penalties, a standard set of dose constrains was found, resulting in a homogeneous dose distribution in the target volumes. The dose to the critical organs was thereby still acceptable. Using these constraints,
Symposia/Proffered papers
dose calculations showed that 7 equi-angular beams were enough to obtain an acceptable dose distribution. For the sequencing procedure 5 to 10 intensity levels were enough to obtain a homogeneous dose distribution. Furthermore it was possible to escalate the dose in the GTV by approximately 10 %, while not significantly increasing the dose to the normal tissues. Conclusion: A class solution for the treatment of 0ropharyngeal cancer has been developed. Using this class solution it is possible to improve the dose distribution compared with the conventional dose distribution and to give an extra boost to the GTV. 74
oral
Forward versus inverse planning of static IMRT of prostate cancer L.J. BO$, R.W. de Boer, B.J. Mijnheer, J.V. Lebesque, E.M.F. Damen The Netherlands Cancer Institute/Antoni van Leeuwenhoek Hospital, Amsterdam, The Netherlands
Purpose: to compare two methods for creating beam segments for static IMRT of prostate cancer. Methods: the comparison was performed for a 5-field prostate treatment using a simultaneous boost technique. The CTV plus a 10mm margin (PTV1) was irradiated to 68 Gy, and a smaller volume, CTV plus a variable margin (0 mm towards the rectum, and 5 mm elsewhere) (PTV 2) was simultaneously irradiated to 78 Gy. Forward planning: beam segments were designed semi-automatically with our planning system (U-MPlan, University of Michigan) using the internal geometry of the patient in beam's eye view. For each beam direction an open conformal segment, and a rectumshielding segment were created for PTV1 and PTV2, respectively. Inverse planning: the KonRad inverse planning system (MRC systems) was used to generate an intensity profile for each beam direction. The intensity profile was obtained iteratively using a cost function. Next, the intensity profile was converted (Radiotherapy Department, University of Utrecht) into a set of segments and imported in our planning system. For both methods beam weight optimisation was subsequently applied. Segments having less than 4 MU were omitted. We evaluated five patients with varying degree of overlap between PTVl and rectum. Dose homogeneity in the target volumes, the dose in the rectal wall and the final number of segments were scored. Results: 3D Dose distributions in PTV1, PTV2 and rectal wall were comparable for both methods of creating beam segments. Forward planning resulted in approximately 20 segments. After beam weight optimisation 7 to 13 segments remained, depending on the overlap between PTV1 and rectum. With inverse planning the number of segments ranged from 35 to 70, and also reduced to approximately 13 segments after optimisation. Conclusion: so far, inverse planning of segments for static IMRT did not improve the 3D dose distribution compared to forward planning of segments. 75
oral
Treatment planning for IMRT using pre-determined intensity modulations for particular target geometries - a feasibility study J.H.L. Mott, G.F. Budge#, P.C. Williams North Western Medical Physics, Christie Hospital NHS Trust, Manchester, UK There is currently much debate over the relative merits of inverse and forward planning approaches to IMRT. Both methods often entail the use of "class solutions" for particular disease sites, in which a number of parameters are predetermined, such as the number and direction of beams and, for inverse methods, the optimisation parameters. This work investigates the possibility of going one stage further and establishing the intensity modulation required to produce the required dose distribution for a particular target geometry, then applying this "generic dose profile" solution directly to patients. If the dose profiles at depth required to produce a particular dose distribution in a standard geometry can be found using a phantom study, they may then be combined with patient specific customised compensation to produce the same dose distribution in individual patients. For a group of patients with similar target geometries this could avoid the need for full inverse planning on every patient and would automatically produce smooth intensity maps that could be delivered with maximum efficiency. Alternatively, the resulting input fluence could be used as the starting point for an
72
Tuesday, 18 September 2001
oral
A treatment planning comparison of 3D conformal therapy, intensity modulated photon therapy and proton therapy for treatment of advanced head and neck tumours L. Cozzi, A. Fogliata, A. Bolsi IOSI, Med. Phys., Bellinzona, CH
Background and Purpose: In this work, the potential benefits and limitations of different treatment techniques, based on mixed photon-electron beams, 3D conformal therapy, intensity modulated photons (IM) and protons (passively scattered and spot scanned), have been assessed using comparative treatment planning methods in a cohort of patients presenting with advanced head and neck tumours. Material and Methods: Plans for five patients were computed for all modalities using CT scans to delineate target volume (PTV) and organs at risk (OAR) and to predict dose distributions. The prescribed dose to the PTV was 54 Gy, whilst the spinal cord was constrained to a maximum dose of 40.5 Gy for all techniques. Dose volume histograms were used for physical and biological evaluation, which included equivalent uniform dose (EUD) calculations. Results: Excluding the mixed photon-electron technique, PTV coverage was within the defined limits for all techniques, with protons providing significantly improved close homogeneity, resulting in correspondingly higher EUD results. For the spinal cord, protons also provided the best sparing with maximum doses as low as 17 Gy. Whilst the IM plans were demonstrated to be significantly superior to non modulated photon plans, they were found to be inferior to protons for both criteria. A similar result was found for the parotid glands. Although they are partially included in the treated volume there is a clear indication that protons, and to a lesser extent IM photons, could play an important role in preserving organ functionality with a consequent improvement of the patient's quality of life. Conclusions: For advanced head and neck tumours, we have demonstrated that the use of IM photons or protons both have the potential to reduce the possibility of spinal cord toxicity. In addition, a substantial reduction of dose to the parotid glands through the use of protons enhances the interest for such a treatment modality in cases of advanced head and neck tumours. However, in terms of target coverage, the use of 3D conformal therapy, although somewhat inferior in quality to protons or IM photons, has been shown to be a reasonable alternative to the more advanced techniques. In contrast, the conventional technique of mixed photon and electron fields has been shown to be inferior to all other techniques for both target coverage and OAR involvement. 73
oral
Intensity modulated radiotherapy for oropharyngeal cancer: a class solution B. van A~selen, H. Dehnad, C.P.J. Raaijmakers, C.H.J. Terhaard, J.J.W. Lagendijk U.M.C. Utrecht, Department of Radiotherapy, Utrecht, The Netherlands
Purpose: Using intensity modulated radiotherapy (IMRT), it might be possible to increase the dose to the gross tumor volume (GTV) of the oropharynx in order to enhance the tumor control probability (TCP). At the same time the normal tissue complication probability should be lower or similar to that obtained for the conventional technique. The aim of this study is to obtain a class solution, i.e. a standardized treatment planning procedure, for the IMRT treatment of the oropharynx. Methods: Ten patients underwent a planning CT scan. The clinical target volume (CTV) of the lymph nodes, the CTV and GTV of the primary target and the critical organs (myelum, brain and parotid glands) were delineated. A 5 mm margin was added to the CTV to define the planning target volume (PTV). The IMRT plans were generated using an inverse planning system. Sequencing was done in a 'step and shoot' mode using a sequencer developed at our institute. After sequencing, the dose calculations for IMRT plans were performed using a 3D calculation algorithm, which was also used to for the conventional plans. In order to find the class solution, the number of beams, the number of intensity levels and the optimal set of dose constraints (penalty settings) were investigated. Results: Preliminary results show that it is possible to obtain a class solution for the IMRT treatment of the oropharynx tumors. By systematically varying the penalties, a standard set of dose constrains was found, resulting in a homogeneous dose distribution in the target volumes. The dose to the critical organs was thereby still acceptable. Using these constraints,
Symposia/Proffered papers
dose calculations showed that 7 equi-angular beams were enough to obtain an acceptable dose distribution. For the sequencing procedure 5 to 10 intensity levels were enough to obtain a homogeneous dose distribution. Furthermore it was possible to escalate the dose in the GTV by approximately 10 %, while not significantly increasing the dose to the normal tissues. Conclusion: A class solution for the treatment of 0ropharyngeal cancer has been developed. Using this class solution it is possible to improve the dose distribution compared with the conventional dose distribution and to give an extra boost to the GTV. 74
oral
Forward versus inverse planning of static IMRT of prostate cancer L.J. BO$, R.W. de Boer, B.J. Mijnheer, J.V. Lebesque, E.M.F. Damen The Netherlands Cancer Institute/Antoni van Leeuwenhoek Hospital, Amsterdam, The Netherlands
Purpose: to compare two methods for creating beam segments for static IMRT of prostate cancer. Methods: the comparison was performed for a 5-field prostate treatment using a simultaneous boost technique. The CTV plus a 10mm margin (PTV1) was irradiated to 68 Gy, and a smaller volume, CTV plus a variable margin (0 mm towards the rectum, and 5 mm elsewhere) (PTV 2) was simultaneously irradiated to 78 Gy. Forward planning: beam segments were designed semi-automatically with our planning system (U-MPlan, University of Michigan) using the internal geometry of the patient in beam's eye view. For each beam direction an open conformal segment, and a rectumshielding segment were created for PTV1 and PTV2, respectively. Inverse planning: the KonRad inverse planning system (MRC systems) was used to generate an intensity profile for each beam direction. The intensity profile was obtained iteratively using a cost function. Next, the intensity profile was converted (Radiotherapy Department, University of Utrecht) into a set of segments and imported in our planning system. For both methods beam weight optimisation was subsequently applied. Segments having less than 4 MU were omitted. We evaluated five patients with varying degree of overlap between PTVl and rectum. Dose homogeneity in the target volumes, the dose in the rectal wall and the final number of segments were scored. Results: 3D Dose distributions in PTV1, PTV2 and rectal wall were comparable for both methods of creating beam segments. Forward planning resulted in approximately 20 segments. After beam weight optimisation 7 to 13 segments remained, depending on the overlap between PTV1 and rectum. With inverse planning the number of segments ranged from 35 to 70, and also reduced to approximately 13 segments after optimisation. Conclusion: so far, inverse planning of segments for static IMRT did not improve the 3D dose distribution compared to forward planning of segments. 75
oral
Treatment planning for IMRT using pre-determined intensity modulations for particular target geometries - a feasibility study J.H.L. Mott, G.F. Budge#, P.C. Williams North Western Medical Physics, Christie Hospital NHS Trust, Manchester, UK There is currently much debate over the relative merits of inverse and forward planning approaches to IMRT. Both methods often entail the use of "class solutions" for particular disease sites, in which a number of parameters are predetermined, such as the number and direction of beams and, for inverse methods, the optimisation parameters. This work investigates the possibility of going one stage further and establishing the intensity modulation required to produce the required dose distribution for a particular target geometry, then applying this "generic dose profile" solution directly to patients. If the dose profiles at depth required to produce a particular dose distribution in a standard geometry can be found using a phantom study, they may then be combined with patient specific customised compensation to produce the same dose distribution in individual patients. For a group of patients with similar target geometries this could avoid the need for full inverse planning on every patient and would automatically produce smooth intensity maps that could be delivered with maximum efficiency. Alternatively, the resulting input fluence could be used as the starting point for an