Motion compensation with a scanned ion beam: a technical feasibility study
- 5.4k Downloads
Intrafractional motion results in local over- and under-dosage in particle therapy with a scanned beam. Scanned beam delivery offers the possibility to compensate target motion by tracking with the treatment beam.
Lateral motion components were compensated directly with the beam scanning system by adapting nominal beam positions according to the target motion. Longitudinal motion compensation to mitigate motion induced range changes was performed with a dedicated wedge system that adjusts effective particle energies at isocenter.
Lateral compensation performance was better than 1% for a homogeneous dose distribution when comparing irradiations of a stationary radiographic film and a moving film using motion compensation. The accuracy of longitudinal range compensation was well below 1 mm.
Motion compensation with scanned particle beams is technically feasible with high precision.
KeywordsDose Distribution Target Motion Motion Compensation Beam Position Pencil Beam
In conformal radiotherapy, geometric margins are commonly used to account for intra-fractional target motion [1, 2]. These margins inevitably lead to inclusion of healthy tissue in the treated volume. In intensity modulated radiotherapy, additional motion effects arise due to so called interplay effects [3, 4, 5]. Treatments are delivered in small partial doses that only result in adequate total dosage if they match as intended. In anatomy's eye view, target motion leads to relative displacement of partial dose depositions and therefore results in local over- and under-dosage.
In a pilot project at Gesellschaft für Schwerionenforschung (GSI) [6, 7, 8, 9], approximately 400 patients have been treated with scanned carbon ion beams with the rasterscan system . For raster scanning, the target volume is divided in slices corresponding to equal ion energies. Irradiations are performed slice-by-slice. The required particle energy is requested from the synchrotron for each slice. Within each slice, a narrow pencil beam is scanned on a virtual raster grid. To achieve the desired dose distribution, the number of particles is optimized for each raster position during treatment planning including biological effects [11, 12, 13, 14, 15, 16]. The scanning progress is intensity controlled. The carbon ion pencil beam is directed to the next raster position by a magnetic deflection system as soon as the planned number of particles has been deposited. After all points within a slice have been irradiated, the beam is aborted and the next energy level is requested from the accelerator. To date, only patients with tumors that are not subject to intra-fractional motion have been treated [7, 17, 18, 19]. For treatments with scanned particle beams, target motion would inevitably lead to local over- and under-dosage due to the relative lateral motion between pencil beam positions as well as possible motion induced changes in radiological depths.
To treat moving targets, while maintaining the conformity between target and treated volume as well as avoiding local over- and under-dosage, we are investigating and developing a system to adapt 3D pencil beam positions to actual target positions in real time. Initially, simulation studies were performed to investigate the potential of target tracking with a scanned ion beam [4, 20]. In beam's eye view, lateral motion adaptation of pencil beam positions is feasible by applying offsets to the raster scanner settings. Real time energy adaptation to compensate changes in radiological depth with the synchrotron directly is not (yet) possible. Therefore online adaptation of particle ranges has to be performed with an additional, dedicated energy modulation system. One of the possibilities is to use a dedicated absorber wedge system .
Prototype systems for lateral as well as longitudinal target tracking with a scanned ion beam have been developed. Experimental results are presented to demonstrate the feasibility of target tracking with a scanned ion beam and to show the performance of the individual prototype tracking sub-systems.
Simulation of target motion
Lateral target motion orthogonal to the beam direction was achieved with a three-axes positioning table. A radiographic film was mounted on the table as detector. The motion was sinusoidal with a period of ~10 s and amplitudes of ± 15 mm in horizontal as well as vertical direction. No external motion monitoring device was used, instead table motion was continuously measured with encoders. Target displacements were evaluated from encoder data and sent directly to the therapy control system (TCS) for beam adaptation during irradiations.
To simulate motion induced variations in particle range, different particle energies were requested from the synchrotron. In a first experiment, three different particle energies were requested from the accelerator repeatedly in fixed order. The energy modulation system was used to adapt the effective particle energy at isocenter to the middle energy. In a second experiment, six different particle energies were requested in mixed order to test the functionality of the system for variable and alternating energy modulations. The maximum difference in energy corresponded to a water equivalent range difference of 27 mm. Again, the energy modulation system was used to adapt the effective particle energy to a single range.
3D online motion compensation
Lateral motion compensation
The raster scanning process is controlled by the TCS. Beam position as well as delivered number of particles are monitored in intervals of ~150 μs and ~10 μs respectively. The standard TCS can adjust small deviations of the actual beam position via a fast feedback loop. Whenever the beam position has been measured, possible deviations are fed back to the control of the scanning magnets to correct the beam position to the nominal position. Typically, deviations are within ± 0.5 mm and corrected after each measurement cycle. The irradiation time for an individual raster point is typically in the order of 5–10 ms.
Several processes are running simultaneously in the TCS including monitoring of the beam intensity, the beam position, and the raster scanner magnet settings. The individual processes communicate via a control loop as well as shared memory. For motion compensation, adaptation of lateral pencil beam positions was implemented by dynamically changing the nominal values of the beam positions in shared memory. As soon as the nominal values have been changed, the feedback loop adjusts the beam position accordingly. A dedicated, additional process running on the TCS receives displacement vectors and then changes the nominal beam positions in shared memory accordingly. In order to avoid hardware changes within the TCS for the prototype setup, a standard network connection (100 Hz) was used to transmit displacement vectors to the TCS. The actual displacement vector is added to the stationary nominal raster point position to compute the new, dynamic nominal position.
Longitudinal motion compensation
Measurement and analysis of dose distributions
Different detectors were used to measure dose distributions: planar radiographic films for lateral 2D dose distributions and a range telescope for longitudinal 1D depth dose distributions .
with D i dose to each individual pixel, N number of pixels within the target area, and mean dose within the target area.
The range telescope was used to measure depth dose distributions, so called Bragg peaks. The telescope consists of two parallel plate ionization chambers in front of and behind a water tank of variable thickness [22, 25, 26]. During the measurements, the thickness of the water tank was increased in steps of 50 μm.
Lateral motion compensation
Longitudinal motion compensation
The results of our feasibility study demonstrate that motion compensation with scanned particle beams is feasible with high precision. Lateral as well as longitudinal compensation were successfully performed during irradiations. In a next step, both motion compensation sub-systems have to be integrated in the therapy control system. Especially replacing standard network connections to transmit compensation parameters should improve the reliability of the system. Furthermore, hardware improvements of the energy modulation system for longitudinal range compensation should be investigated, and implementation of motion monitoring has to be developed.
Re-design of the wedge system for fast longitudinal motion compensation is advisable since the thickness of the wedges can most likely be reduced to the compensation range required for patient treatments in order to reduce lateral scattering as well as fragmentation of the primary particle beam [27, 28, 29]. Furthermore, the active area of the wedge system (120 × 150 mm2) does currently not match the treatment area of the scanning system (200 × 200 mm2). The wedge size thus has to be increased at least in vertical direction. In contrast, the horizontal dimension of the active area does not necessarily have to match the scan area. If the center of mass of the wedges follows the left-right motion of the ion beam during raster scanning, an active area that is smaller than the maximum treatment area is sufficient. However, less wedge motion and therefore reduced system performance is required if the active area is sufficiently large to cover the complete scanning area. Detailed requirements on the compensation speed have to be derived from simulation studies, for example based on 4D computed tomography data [30, 31, 32].
Another problem of motion tracking that has not yet been solved adequately is precise monitoring of target motion. To date, several different methods have been reported in the literature. Currently, the most promising technique seems to be fluoroscopic motion detection because target motion is imaged directly [33, 34, 35, 36, 37, 38]. Other techniques that monitor external surface motion have to be evaluated regarding the accuracy to derive target positions [39, 40, 41, 42, 43, 44, 45, 46]. Since the particle range and thus the Bragg peak position are influenced by target motion and currently no motion monitoring system exists to determine changes in water-equivalent range a link to 4D treatment planning is required [47, 48]. Motion states from 4DCT which are used to determine range changes could be detected by motion monitoring. Compensation vectors are then calculated during treatment planning and applied according to detected motion states. In case of motion irregularities or unknown motion states the treatment can be paused until the patient is back to normal breathing.
The results of our study demonstrate the high precision that is technically feasible for motion tracking with scanned particle beams. Lateral motion compensation restored homogeneous dose distributions delivered to moving targets. Differences in dose uniformity between irradiation of a stationary radiographic film and a moving film using motion compensation were below 1%. Longitudinal compensation precision was well below 1 mm.
- 1.ICRU: Report 50. Bethesda, Md, USA, International Commission on Radiation Units and Measurements; 1993.Google Scholar
- 2.ICRU: Report 62. Bethesda, Md, USA, International Commission on Radiation Units and Measurements; 1999.Google Scholar
- 14.Scholz M, Kraft G: Calculation of heavy ion inactivation probabilities based on track structure, x ray sensitivity and target size. Radiat Prot Dosim 1994, 52: 29-33.Google Scholar
- 19.Nikoghosyan A, Schulz-Ertner D, Didinger B, Jakel O, Zuna I, Hoss A, Wannenmacher M, Debus J: Evaluation of therapeutic potential of heavy ion therapy for patients with locally advanced prostate cancer. Int J Radiat Oncol Biol Phys 2004, 58: 89-97. 10.1016/S0360-3016(03)01439-1CrossRefPubMedGoogle Scholar
- 22.Sihver L, Schardt D, Kanai T: Depth-dose distributions of high-energy carbon, oxygen and neon beams in water. Jpn J Med Phys 1998, 18: 1-21.Google Scholar
- 23.Bathelt B: Filmdosimetrie in der Schwerionen-Tumortherapie: 3-dimensionale Dosisverifikation in gemischten Teilchenstrahlfeldern. Gesamthochschule Kassel; 2000.Google Scholar
- 25.Schardt D, Stelzer H, Junk H, Arndt U: Bragg curve measurements with ionisation chambers. Grundinger, U. 336. Darmstadt, Gesellschaft für Schwerionenforschung mbH. GSI Scientific Report 1992 1993.Google Scholar
- 28.Schall I, Schardt D, Geissel H, Irnich H, Kankeleit E, Kraft G, Magel A, Mohar MF, Mnnzenberg G, Nickel F, Scheidenberger C, Schwab W: Charge-changing nuclear reactions of relativistic light-ion beams (5 <= Z <= 10) passing through thick absorbers. Nucl Instrum Meth B 1996, 117: 221-234. 10.1016/0168-583X(96)00325-4CrossRefGoogle Scholar
- 29.Gunzert-Marx K, Iwase H, Schardt D, Simon RS: Secondary beam fragments produced by 200 MeVu(-1) C-12 ions in water and their dose contributions in carbon ion radiotherapy. New Journal of Physics 2008., 10:Google Scholar
- 34.Berbeco RI, Jiang SB, Sharp GC, Chen GTY, Mostafavi H, Shirato H: Integrated radiotherapy imaging system (IRIS): design considerations of tumour tracking with linac gantry-mounted diagnostic x-ray systems with flat-panel detectors. Phys Med Biol 2004, 49: 243-255. 10.1088/0031-9155/49/2/005CrossRefPubMedGoogle Scholar
- 35.Shirato H, Harada T, Harabayashi T, Hida K, Endo H, Kitamura K, Onimaru R, Yamazaki K, Kurauchi N, Shimizu T, Shinohara N, Matsushita M, aka-Akita H, Miyasaka K: Feasibility of insertion/implantation of 2.0-mm-diameter gold internal fiducial markers for precise setup and real-time tumor tracking in radiotherapy. Int J Radiat Oncol Biol Phys 2003, 56: 240-247.CrossRefPubMedGoogle Scholar
- 36.Shimizu S, Shirato H, Kitamura K, Shinohara N, Harabayashi T, Tsukamoto T, Koyanagi T, Miyasaka K: Use of an implanted marker and real-time tracking of the marker for the positioning of prostate and bladder cancers. Int J Radiat Oncol 2000, 48: 1591-1597. 10.1016/S0360-3016(00)00809-9CrossRefGoogle Scholar
- 46.Willoughby TR, Forbes AR, Buchholz D, Langen KM, Wagner TH, Zeidan OA, Kupelian PA, Meeks SL: Evaluation of an infrared camera and X-ray system using implanted fiducials in patients with lung tumors for gated radiation therapy. Int J Radiat Oncol Biol Phys 2006, 66: 568-575.CrossRefPubMedGoogle Scholar
- 48.Bert C, Rietzel E: 4D treatment planning for scanned ion beams. Radiat Oncol 2007., 2:Google Scholar