Technology in Cancer Research and Treatment ISSN 1533-0346 Volume 14 Number 5 October 2015 2014 June 16. Epub ahead of print.

Effect of Scanning Beam for Superficial Dose in Proton Therapy www.tcrt.org DOI: 10.7785/tcrt.2012.500435 Proton beam delivery technology is under development to minimize the scanning spot size for uniform dose to target, but it is also known that the superficial dose could be as high as the dose at Bragg peak for narrow and small proton beams. The objective of this study is to explore the characteristics of dose distribution at shallow depths using Monte Carlo simulation with the FLUKA code for uniform scanning (US) and discrete spot scanning (DSS) proton beams. The results show that the superficial dose for DSS is relatively high compared to US. Additionally, DSS delivers a highly heterogeneous dose to the irradiated surface for comparable doses at Bragg peak. Our simulation shows that the superficial dose can become as high as the Bragg peak when the diameter of the proton beam is reduced. This may compromise the advantage of proton beam therapy for sparing normal tissue, making skin dose a limiting factor for the clinical use of DSS. Finally, the clinical advantage of DSS may not be essential for treating uniform dose across a large target, as in craniospinal irradiation (CSI).

Vadim P. Moskvin, Ph.D.* Neil C. Estabrook, M.D.* Chee-Wai Cheng, Ph.D. Indra J. Das, Ph.D. Peter A. S. Johnstone, M.D. Department of Radiation Oncology, Indiana University School of Medicine, Indianapolis, IN, USA IU Health Proton Therapy Center, Bloomington, IN, USA 535 Barnhill Dr., RT-041, Indianapolis, IN 46202, USA

Key words: FLUKA; Monte Carlo simulation; Proton beam therapy; Scanning proton beam; Skin toxicity; Spot scanning; Uniform scanning.

Introduction Advances in proton therapy delivery methods have sprouted debate over the strengths and weaknesses of different systems. The earliest design and most popular system employed a passive scattering technique in which a proton beam passes through scattering foils and propellers or ridge filters (1) to create a spread out Bragg peak (SOBP). In this design beam shaping is typically performed with patient specific devices such as apertures and compensators. However an alternative beam delivery design available to some proton treatment centers is to actively scan the proton beam to treat the desired volume within the patient. It is widely believed that the newer active scanning techniques have advantages over passive scanning in their ability to create more conformal dose distributions but more importantly to decrease both the proton beam neutron contamination and its level of induced radioactivity in the field shaping apertures (2-4). Although there are multiple technologies now employed for actively scanning a proton beam, two main types are discussed here: uniform scanning (US) and discrete spot scanning (DSS). In an US system, the proton beam is directed using scanning magnets and a range modulator to create a uniform field and SOBP at Abbreviations: US: Uniform Scanning; DSS: Discrete Spot Scanning; PDD: Percent Depth Doses; FWHM: Full Width at Half Maximum; SOBP: Spread Out Bragg Peak; CSI: Cranial Spinal Irradiation; IMPT: Intensity Modulated Proton Therapy; ICBM: Ion Chamber Beam Monitor; PDM: Patient Dose Monitor.

*Corresponding authors: Vadim P. Moskvin, Ph.D. Neil C. Estabrook, M.D. E-mail: [email protected] (Vadim P. Moskvin); [email protected] (Neil C. Estabrook)

643 Downloaded from tct.sagepub.com at UNIV OF LETHBRIDGE on November 10, 2015



Moskvin et al.

depth as shown in several publications (5-7). For this type of system, the beam intensity remains constant and is swept over the transverse area of the patient’s treatment field (with pre-specified additional margins) in contiguous end-to-end and layer-by-layer passes of scan lines using range modulators. Distinctly, DSS systems are able to deliver the proton beam to a discrete spot within the treatment field, with the beam cycled off while it is directed to the next area of the field before it is turned on again. Moreover, the beam intensity may vary as the beam is swept through the treatment field (without additional margins) layer-by-layer. Therefore, as its name would suggest, a uniform dose is delivered with the US method, while DSS results in a number of discretely placed overlapping “spots” in the treatment field with possibly uneven dose. It is a general community belief that DSS is superior to US because of its potential ability for better conformation of a high dose region to the target volume and potential for intensity modulated proton therapy (IMPT). However, it should be pointed out that even the smallest achievable diameter of a scanning beam is not capable of delivering an optimum dose distribution at every spatially located target volume (8). With currently available beam shaping technology, beam diameters as narrow as 1.2 cm full width half maximum (FWHM) are achievable and state-of-the-art proton delivery systems dedicated for spot-scanning have demonstrated that beam diameters vary from 1.2 cm at 221.8 MeV to 3.4 cm at 72.5 MeV in air (6, 9). It has been shown recently that spot sizes (FWHM) of 1.1 cm for 140 MeV and 0.68 cm for 226.7 MeV in air are realistically achievable with the novel IBA pencil beam system (10). The progress in development of the laser driven proton accelerator technology can bring the spot size to few mm (11).

The Hitachi delivery system contains helium chambers at the beam pass (6, 7, 9). To minimize beam scattering in air, a vacuum inside the nozzle is assumed. The simulation of a proton beam passing through the nozzle starts after the Quadrupole Magnets I. The model layout included an ion chamber beam monitor (ICBM), located upstream of the scanning magnet, and a patient dose monitor (PDM) inside the nozzle (see Figure 1A). Nozzles usually have two scanning magnets X and Y, however, a single magnet design is available (15) and was modeled here. The ICBM and PDM were simulated as those used at the IU Health Proton Therapy Center (5, 7). This layout is called a “conventional nozzle” layout in this paper. One possible modification of the layout is modeled reflecting the recent trend in nozzle design from IBA (7, 10) which includes Quadrupole Magnets II set after the first monitor (ICBM) and moving the second monitor (PDM) to the exit of the nozzle (see Figure 1B). The nozzle contains a vacuum chamber in the beam path. This layout is called a “modified nozzle” or “pencil beam nozzle” layout in this paper. The air gap between the nozzle and a water phantom is set to 5 cm for all simulations. The mechanical isocenter is placed at 32 cm from the nozzle exit. Proton Beam Characterization The characteristics of the beam entering the nozzle were set to mimic currently achievable beam parameters (7, 9, 10, 16).

Previous models for creating proton microbeams (10 mm diameter) for radiosurgery have shown that surface dose can increase to levels far above the Bragg peak dose as the beam diameter is made smaller (12-14). The potential benefits of IMPT using DSS need to be weighed against the possibility of unintentional adverse effects, such as overdosing superficial normal tissues at the entrance area of the beam. The objective of this study is to ascertain and compare the differences in dose at shallow depths using Monte Carlo simulation of the two different proton beam scanning techniques, US and DSS. Methods and Materials Beam Nozzle Layout Models The scanning beam nozzle layout model generalizes a design including typical components (see Figure 1A). This nozzle layout mimics an IBA Universal Nozzle which contains air.

Figure 1:  Layouts of proton treatment nozzles used in Monte Carlo simulation. (A) Conventional nozzle layout. L1   270.5 cm, L2    231.3 cm, and L3  168.9 cm and (B) modified nozzle layout. Range modulator plates are off beam corresponding to a position of zero pristine Bragg peak pull back. The air gap L4   5 cm.

Technology in Cancer Research & Treatment 2014 June 16. Epub ahead of print Downloaded from tct.sagepub.com at UNIV OF LETHBRIDGE on November 10, 2015

Superficial Dose with Proton Scanning

645

The lateral proton fluence of a beam entering the nozzle is modeled as a Gaussian distribution with σ of 0.1 cm for all energies in the conventional nozzle layout. The influence of the beam size reduction on spot size is simulated with a smaller σ for illustration. A set of the sizes ranging from σ of 0.021 to 0.637 cm (FWHM of 0.05 to 1.5 cm) are simulated for the beam output of the modified nozzle layout. The energy spectra (energy spread) of protons is approximated by a Gaussian distribution. Beam energies for the simulation are chosen to be 93.5, 130.8, 152.6, 173.9 and 206.8 MeV with an energy spread (FWHM ) of 0.8, 2.2, 1.4, 1.3 and 2.1% respectively. The angular spread of the beam is set at 0.003 mrad.

Figure 2:  Scanning pattern for (A) US and (B) DSS.

Monte Carlo Simulation The Monte Carlo simulation is performed to generate three dimensional dose distributions in air and water phantom for a single spot from the proton beam in the two nozzle layouts described above. Monte Carlo general-purpose particle transport code FLUKA version 2011.2.16 is used in this simulation (17, 18). The default setting ‘PRECISIO’ in FLUKA general-­purpose Monte Carlo code is applied to customize the settings of the physical model used in the simulation (18). The ­initial proton transport was simulated with a cutoff energy at 100 keV. The delta-electron transport is included in this simulation. The USRBIN cards are used for scoring the ­ ­proton dose. Twenty million initial protons are used to generate the dose from the scanning beam spot in 3D space with a voxel size of 0.010.010.05 mm. The estimated ­statistical error was not more than 2% along the beam axis. The Monte Carlo generated data sets for single spots are used for numerical simulation of the scanning pattern. The spot characteristics in air, in water phantom, and dose deposition in the two different scanning scenarios, namely US and DSS, are simulated. The numerical integrations of US and DSS beam patterns (see Figure 2) were performed with the aid of MATLAB (R2012a) (The MathWorks, Inc.) program (19). The beam spreads in both air and water phantoms are simulated to characterize energy deposition from modeled scanning beams. The beam spread is described by the parameter sigma1 (σ), the Gaussian approximation of the main part of the dose distribution in a single spot (20). The results are presented below in terms of σ. The parameter σ is mathematically related to full width half maximum (FWHM) by FWHM  2.35482 σ.

1

The offset Δ (see Figure 2) between scanning lines in a US method or between spots in a DSS method is set to 2.2-2.3 of beam spread σair in air. This offset Δ is the mathematical optimum summation of two Gaussian distributions to achieve a satisfactory uniformity in dose coverage while keeping the lateral dose edges reasonably sharp (21). Two iterative scanning scenarios are considered to compare doses at shallow depths for US and DSS active proton beam delivery techniques when treating hypothetical small fields of both 0.9   0.9 cm2 and 1 3  12 cm2, as well as a large 7 3  12 cm2 field. The offset Δ center-to-center of the discrete spot shots for the DSS models are 2.3 σair for a single shot in air (see Figure 2). US is performed with the same 2.3 σair offset between scanning lines. The SOBP was calculated for the scanning beam in the modified nozzle layout applying the range modulator plates modeled as layers of polymethylmethacrylate (PMMA) (density of 1.157 g/cm3) with the thicknesses assuring a 3 mm pull back of the pristine Bragg peak for each layer. The 3 cm SOBP for the fields 0.9 3  0.9 cm2 and 3 × 3 cm2 are simulated for the DSS method. The dose fields for SOBPs are calculated with the reference to σwater in water phantom in the position of the distal spots. Results Proton Beam Spot Size in Air and Water for Conventional Nozzle Layout The results of the simulation of the spot size in air and in water phantom for beam parameters currently achievable in a conventional nozzle layout are provided in Table I. The beam spot is usually characterized by the size of the beam in air at isocenter. It is seen from the Table I that actual sizes of the spot at the position of the Bragg peak in a water phantom differs significantly from those stated in air for the therapeutic

Technology in Cancer Research & Treatment 2014 June 16. Epub ahead of print Downloaded from tct.sagepub.com at UNIV OF LETHBRIDGE on November 10, 2015



Moskvin et al. Table I Proton beam spread in air and water.

Beam energy (MeV)

Spot size in air at the surface of the water phantom σph (cm)

Spot size in air at the isocenter σair (cm)

Spot size in water at the position of Bragg peak σwater (cm)

130.9 152.6 173.9 206.8

0.414 0.352 0.314 0.269

0.464 0.390 0.365 0.296

0.535 0.543 0.589 0.696

beam energies studied (206.8 to 130.9 MeV). The beam spot size at the isocenter in air has a Gaussian distribution of particle fluence with a sigma ranging from σair 5 0.3 to 0.5 cm for high and low energies, respectively. However, the beam spot size in water, σwater is about 0.5 to 0.7 cm at the position of the Bragg peak for these therapeutic proton energies. The reduction of the beam size at the nozzle entrance does not affect the size of the spot considerably for a conventional nozzle layout. Thus, the beam size of σ 5 0.1 cm at the entrance

Figure 3:  The dose distribution and PDD from a spot proton beam of σ 5 0.1 cm at the entrance of the nozzle in water phantom.

Technology in Cancer Research & Treatment 2014 June 16. Epub ahead of print Downloaded from tct.sagepub.com at UNIV OF LETHBRIDGE on November 10, 2015

Superficial Dose with Proton Scanning of the nozzle produces a spot size in air of σair 5  0.39 cm at the isocenter for a beam with a 16 cm range. Reductions of the beam size to σ 5 0.025 and 0.001 cm yields spot sizes σair of 0.3894 and 0.3893 cm at isocenter, respectively. Our model shows that the invariance of the spot size is defined by the scattering of protons on the ICBM and PDM foils and the distance from these monitors to the isocenter forms a source of the scattered proton component for the proton pencil beam. Hence, by minimizing the L3 value in the model (Figure 1) and by compensating for scatter at the ICBM with the addition of Quadrupole Magnets II in the modified nozzle layout, improvements in the achievable beam spot size are obtained. Dose Distribution in Water Phantom for a Single Spot Proton Beam The dose distributions and percentage depth dose (PDD) of a single spot proton beam with the width σ 5 0.1 cm at the entrance of the nozzle for a conventional nozzle layout are

647 shown in the Figure 3 for the three proton ranges of 6.5, 16 and 27 cm in water phantom. As shown in Figure 3, the dose distribution forms a maximum at the shallow depths with an increase of the initial beam energy. The relative superficial dose is higher for higher energy beam. For example the superficial dose is 160% of the Bragg Peak dose for the 27 cm range beam for small field of (a beam with σ 5 0.1 cm at the entrance of the nozzle). Furthermore, Figure 4A illustrates that there is dose enhancement at shallow depths that increases with the reduction of the incident beam size. These surface doses were calculated using the modified nozzle layout. In this case, some of the beams we modeled are smaller than what is currently achievable with known technology, specifically those proton beams with σair smaller than 0.3 cm. Our data also show that reducing beam spot size in air does not reduce spot size at the Bragg peak. As seen from Figure 4B, there is a limit of our ability to reduce the spot size in water as the spot size is air is made narrower. For example, the reduction of the spot size

Figure 4:  Single proton spot beam in water: (A) PDDs with various σair for proton beam of 16 cm range in water; (B) spot size in water at the position of Bragg peak σwater for various spot sizes of the beam at the surface of the water phantom σair placed at 5 cm air gap from the nozzle exit; and (C) ratio of the dose at 1 mm depth in water phantom to the dose at the Brag peak for various spot sizes of the beam at the surface of the water phantom σair.

Technology in Cancer Research & Treatment 2014 June 16. Epub ahead of print Downloaded from tct.sagepub.com at UNIV OF LETHBRIDGE on November 10, 2015



Moskvin et al.

from σair 5 0.085 to 0.0212 cm will reduce a spot in a water phantom from σwater 5 0.369 to 0.355 cm at the position of the Bragg peak for a 16 cm range proton beam. The functional dependence between σair and σwater reaches extremum for smaller spot size (Figure 4B). Figure 4C demonstrates the surface dose theoretically will be ten times higher than Bragg peak dose corresponding to reductions in the beam spot size. This curve reaches an asymptotic value beyond σair of 0.8 cm where superficial dose is in the clinically accepted range of 40-60%.

beam sizes. Figure 5 shows PDDs for both US and DSS methods using a single scan line simulation of 12 cm total length. Figure 5A shows that for a small field of size 1 3  12 cm2 (spot σair 5  0.296 cm) using US, the superficial dose is 60% of the Bragg peak maximum.

US and DSS: Small Fields: Using a conventional nozzle layout, scanning models are performed with currently achievable

Figure 5B shows the PDD curves for the DSS method. The red line represents the PDD passing through the center of the scanning spot. The blue line represents the PDD between the centers of the spots (the offset is of 1.1 σair from the center of the spot). In this case, there is a considerable dose difference between the center of the beam spot shots on the surface and the area between the spots. The difference between two superficial dose PDDs is approximately 18% with the area directly hit by the spot approximately 70% of the Bragg peak dose and the mid-point of the area not

Figure 5:  PDD for a single line scan by 27 cm range proton beam (spot σair = 0.296 cm). (A) Uniform scanning and (B) spot-scanning. Red line represents PDD along the normal to the surface passing through the center of the scanning spot step. Blue line represents PDD along the normal to the surface between the centers of the spots (the offset 1.1 σair).

Figure 6:  Dose distribution for spot-scanning proton beam (spot size σair = 0.0212 cm) covering 0.9 3  0.9 cm2 field. (A) Dose distribution in X,Z plane and (B) lateral in X,Y plane, dose distribution at 1 mm depth.

Dose Distribution in Water Phantom for Scanning Proton Beams

Technology in Cancer Research & Treatment 2014 June 16. Epub ahead of print Downloaded from tct.sagepub.com at UNIV OF LETHBRIDGE on November 10, 2015

Superficial Dose with Proton Scanning

649

Figure 7:  Dose distribution for large fields for US and DSS methods for 27 cm proton beam with σair = 0.296 cm. (A) US – PDDs through the center of the spot (red line), between the lines (blue line); (B) US – lateral dose profile at cross scanning line direction at 1 mm depth; (C) US – lateral dose at 27 mm depth; (D) DSS – PDDs through the center of the spot (red line), between the lines (blue dotted line) and between the lines with offset 1.1 σair (green dashed line); (E) DSS – lateral dose profile at cross spot steps (red line) and between the lines with offset 1.1 σair (blue dotted line) at 1 mm depth in water; (F) DSS – lateral dose at 27 mm depth.

directly in the beam’s path approximately 52% of the Bragg peak dose. Figure 6 illustrates the dose deposition of a theoretical spot size σair 5 0.0212 cm using a DSS technique. The dose distributions from this model show considerable increases in dose at shallow depths up to 2 cm deep. Our data show that the superficial dose in DSS could reach 100% of dose at the Bragg peak with dose variations along the spot’s pathway with a spot step offset of 1.8 σair. US and DSS: Large Fields: The PDDs and lateral beam ­profiles for both US and DSS beam control methods in a conventional nozzle layout are presented in Figure 7 for a larger field, 7 3  12 cm2. The dose distributions are normalized to Bragg peak value. The dose distribution at treatment depth (defined by position of Bragg peak for a given proton energy), is relatively uniform for both US and DSS techniques (Figure 7C and 7F).

Figure 7A, 7B, 7D, and 7E shows that the superficial dose variations are approximately 5% between scan lines at surface depths in the US method and up to 10% in the DSS method. SOBP in a Discrete Spot Scanning Beam The dose distributions for DSS beam control methods in a modified nozzle layout are presented in Figure 8 for 3 3  3 cm2 field. The first layer is defined by the proton beam of 16 cm range. The dose distributions are normalized to the center of the SOBP at a depth of 14.5 cm. Figure 8B presents the dose distribution in the lateral plane (X,Y) at 1 mm depth. Figure 8 shows that the superficial dose variations are approximately 35% between scan spots at surface depths in the DSS method. At 1 mm depth the dose can reach 65% of the dose at the center of SOBP for 3 3  3 cm2 field. As seen in Figures 8 (as well as in Figures 6 and 7) that a “saw tooth” profile is observed up to 3 to 4 cm depth in the water phantom.

Technology in Cancer Research & Treatment 2014 June 16. Epub ahead of print Downloaded from tct.sagepub.com at UNIV OF LETHBRIDGE on November 10, 2015



Moskvin et al. the beam alternates between on and off as it is moved from position to position, the overall profile demonstrates peaks in intensity for each position in the linear field. The result is a “saw tooth” beam profile, as observed up to 3 to 4 cm depths in a water phantom, in a linear scan, and a double saw tooth pattern of considerable dose heterogeneity within a large scanned field (Figure 7). This finding has also been noted in a recent study on proton microbeam applications at 3 and 5 cm depths (14). The degree of dose uniformity at shallower depths for US is due to a summation of dose contributions from the contiguous nature of the beam. High superficial dose for a single proton beam is caused by protons scattered off-axis from the central beam delivering a dose spread laterally to the pristine Bragg peak. This characteristic is clearly evident in the PDD data for DSS beam. Furthermore, the proton fluence at the central axis of the beam is reduced as the beam diameter is made smaller. Therefore, reductions in beam diameter may contribute to further increase in relative dose at superficial depths beyond those already shown in DSS beam.

Figure 8:  Dose distribution for 3 cm SOBP for 3  3 3 cm2 field for the DSS method for proton beam range of 16 cm with σair = 0.0212 cm. (A) Dose distribution in XZ plane and (B) the dose distribution in XY plane at 1 mm depth in water.

Discussion Monte Carlo simulation shows that both US and DSS techniques provide uniform dose coverage at the depth of the pristine Bragg peak. However, the doses at superficial depths have a pronounced heterogeneity in DSS for small fields compared to US. This heterogeneity is less pronounced but not completely eliminated, when comparing the two beam control modes in large scanning fields. The US beam, by its nature of scanning uniformly across the treatment field, has a uniform beam profile while scanning in a contiguous line. There is, however, non-uniform dose at the surface for the spaces in between the parallel scanning lines. The DSS beam has a predictably jagged profile caused by the beam’s toggling between on and off as it is moved into each subsequent position across the irradiated field. Because of the fact that

Our data show that relative superficial dose is much lower than Bragg peak dose when larger beam diameters are used such as σair 5 0.425 and 0. 637 cm that correspond to 1.0 and 1.5 cm diameter FWHM beams respectively. Figure 4 shows that when the diameter of the beam is decreased, the PDDs at superficial depths are correspondingly increased when compared to Bragg peak dose. Even with DSS as an available method for dose painting as a first step toward this level of dose distribution, the beam size based on the present technology is still too broad for IMPT (8). It would seem that as the spot size is further decreased to achieve this goal; the superficial dose increase may become a major issue with small-size proton beam. While our model indicates that it is theoretically possible to decrease the spot size using a modified nozzle layout, it would, on the other hand, call into question the ultimate value of a narrower beam when considering potential increases in superficial dose. The superficial dose variations (up to 70% in the illustrative SOBP case), and the irregularity of its distribution raises the question of the unknown possible effects on healthy tissue. In particular, the experience from photon beam GRID therapy suggest that higher tissue damage may be expected in comparison to a uniform dose field (22). The possibility of the bystander effect (23, 24) can affect the desired goal to minimize the beam size for true voxel-by voxel painting in IMPT with a DSS method. Additional complications could be seen especially for abdominal radiation due to metabolic syndrome (25). This aspect of radiobiological response of healthy tissue has not been previously reported for ion beam therapy. We believe the results of this study indicate that further research is needed into the possibility that DSS and proton microbeams might become mutually exclusive technologies as the

Technology in Cancer Research & Treatment 2014 June 16. Epub ahead of print Downloaded from tct.sagepub.com at UNIV OF LETHBRIDGE on November 10, 2015

Superficial Dose with Proton Scanning skin toxicities might limit their use together. Finally, when treating large targets with uniform dose, such as craniospinal irradiation, there is no advantage for decreasing the size of the beam further than what is currently available. Conclusion Our model for evaluating and quantifying the superficial dose for US versus DSS techniques for controlling a proton beam shows that DSS delivers a highly heterogeneous dose compared to US at superficial depth with comparable doses at the Bragg peak. Superficial dose will become higher as the diameter of the proton beam is narrowed with future developments in nozzle design. The ultimate goal of differentiated proton dose delivery with high resolution in a manner similar to photon intensity modulated radiation therapy may be compromised by the dose heterogeneity at shallower depths. The radiobiological impact to normal tissues of highly heterogeneous doses at superficial depth several times higher than the Bragg peak doses is not known for proton beams. For a single field setup, US with currently achievable beam sizes seems to deliver a more homogenous dose distribution at shallow depths while also maintaining a lower surface dose relative to Bragg peak dose when compared to DSS technique. When comparing the merit of US to DSS, evaluation of dose from scattered particles and neutrons as well as cost associated with both techniques should be carefully examined. Conflict of Interests We certify that regarding this paper, no conflicts of interests exist; the work is original and has not been accepted for publication nor is concurrently under consideration elsewhere. This will not be published elsewhere without the permission of the Editor. All the authors have contributed directly to the planning, execution or analysis of the work reported or to the writing of the paper. References 1. Kostjuchenko, V., Nichiporov, D., Luckjashin, V. A compact ridge filter for spread out Bragg peak production in pulsed proton clinical beams. Med Phys 28, 1427-1430 (2001). DOI: 10.1118/1.1380433 2. Hecksel, D., Anferov, V., Fitzek, M., Shahnazi, K. Influence of beam efficiency through the patient-specific collimator on secondary neutron dose equivalent in double scattering and uniform scanning modes of proton therapy. Med Phys 37, 2910-2917 (2010). DOI: 10.1118/1.3431575 3. Jarlskog, C. Z., Paganetti, H. Risk of developing second cancer from neutron dose in proton therapy as function of field characteristics, organ, and patient age. Int J Radiat Oncol Biol Phys 72, 228-235 (2008). DOI: 10.1016/j.ijrobp.2008.04.069 4. Xu, X. G., Bednarz, B., Paganetti, H. A review of dosimetry studies on external-beam radiation treatment with respect to second cancer induction. Phys Med Biol 53, R193-R241 (2008). DOI: 10.1088/0031-9155/53/13/r01 5. Farr, J. B., Mascia, A. E., Hsi, W. C., Allgower, C. E., Jesseph, F., Schreuder, A. N., Wolanski, M., Nichiporov, D. F., Anferov, V. Clinical characterization of a proton beam continuous uniform

651 s­ canning system with dose layer stacking. Med Phys 35, 4945-4954 (2008). DOI: 10.1118/1.2982248 6. Gillin, M. T., Sahoo, N., Bues, M., Ciangaru, G., Sawakuchi, G., Poenisch, F., Arjomandy, B., Martin, C., Titt, U., Suzuki, K., Smith, A. R., Zhu, X. R. Commissioning of the discrete spot scanning proton beam delivery system at the University of Texas MD Anderson Cancer ­Center, Proton Therapy Center, Houston. Med Phys 37, 154163 (2010). DOI: 10.1118/1.3259742 7. Nichiporov, D., Hsi, W., Farr, J. Beam characteristics in two different proton uniform scanning systems: A side-by-side comparison. Med Phys 39, 2559-2568 (2012). DOI: 10.1118/1.3701774 8. Widesott, L., Lomax, A. J., Schwarz, M. Is there a single spot size and grid for intensity modulated proton therapy? Simulation of head and neck, prostate and mesothelioma cases. Med Phys 39, 1298-1308 (2012). DOI: 10.1118/1.3683640 9. Smith, A., Gillin, M., Bues, M., Zhu, X. R., Suzuki, K., Mohan, R., Woo, S., Lee, A., Komaki, R., Cox, J., Hiramoto, K., Akiyama, H., Ishida, T., Sasaki, T., Matsuda, K. The M. D. Anderson proton therapy system. Med Phys 36, 4068-4083 (2009). DOI: 10.1118/1.3187229 10. Grevillot, L., Bertrand, D., Dessy, F., Freud, N., Sarrut, D. A Monte Carlo pencil beam scanning model for proton treatment plan simulation using GATE/GEANT4. Phys Med Biol 56 (2011). DOI: 10.1088/0031-9155/56/16/008 11. Nishiuchi, M., Sakaki, H., Hori, T., Ogura, K., Yogo, A., Pirozhkov, A. S., Sagisaka, A., Orimo, S., Mori, M., Sugiyama, H., Kiriyama, H., Tampo, M., Daito, I., Okada, H., Kanazawa, S., Tanoue, M., Shimomura, T., Nakai, Y., Sasao, H., Wakai, D., Sasao, F., Bolton, P. R., Daido, H., Kawanishi, S., Iseki, Y., Nagafuchi, T., Maeda, K., Hanawa, K., Yoshiyuki, T., Souda, H., Noda, A. Toward laser driven ­proton medical accelerator. Journal of Physics: Conference Series 244, 022051 (2010). 12. Vatnitsky, S. M., Miller, D. W., Moyers, M. F., Levy, R. P., Schulte, R. W., Slater, J. D., Slater, J. M. Dosimetry techniques for narrow proton beam radiosurgery. Phys Med Biol 44, 2789-2801 (1999). DOI: 10.1088/0031-9155/44/11/308 13. Moyers, M. F., Reder, C. S., Lau, D. C. Generation and characterization of a proton microbeam for experimental radiosurgery. Technol Cancer Res Treat 6, 205-211 (2007). 14. Prezado, Y., Fois, G. R. Proton-minibeam radiation therapy: a proof of concept. Med Phys 40, 031712-031718 (2013). 15. Anferov, V. Combined X-Y scanning magnet for conformal proton radiation therapy. Med Phys 32, 815-818 (2005). DOI: 10.1118/1.1862801 16. Clasie, B., Depauw, N., Fransen, M., Goma, C., Panahandeh, H. R., Seco, J., Flanz, J. B., Kooy, H. M. Golden beam data for proton pencil-beam scanning. Phys Med Biol 57, 1147-1158 (2012). DOI: 10.1088/0031-9155/57/5/1147 17. Battistoni, G., Muraro, S., Sala, P. R., Cerutti, F., Ferrari, A., Roesler, S., Fassò, A., Ranft, J. The FLUKA code: description and benchmarking. Albrow, M., Raja R., (Eds.), Proceedings of the Hadronic Shower Simulation Workshop 2006, Fermilab 6-8 September 2006, AIP Conference Proceedings 896, 31-49 (2007). 18. Ferrari, A., Sala, P. R., Fasso, A., Ranft, J. FLUKA: a multi-particle transport code. CERN 2005-10, INFN/TC_05/11, SLAC-R-773 (2005). 19. MATLAB Release 2012a, The MathWorks, Inc., Natick, Massachusetts, United States (2012). 20. Li, Y. P., Zhu, R. X., Sahoo, N., Anand, A., Zhang, X. D. Beyond Gaussians: a study of single-spot modeling for scanning proton dose calculation. Phys Med Biol 57, 983-997 (2012). DOI: 10.1088/0031-9155/57/4/983 21. Anferov, V. A. Scan pattern optimization for uniform proton beam scanning. Med Phys 36, 3560-3567 (2009). DOI: 10.1118/1.3158731 22. Zwicker, R. D., Meigooni, A., Mohiuddin, M. T ­ herapeutic advantage of grid irradiation for large single fractions. Int J Radiat Oncol Biol Phys 58, 1309-1315 (2004). DOI: 10.1016/j.ijrobp.2003.07.003

Technology in Cancer Research & Treatment 2014 June 16. Epub ahead of print Downloaded from tct.sagepub.com at UNIV OF LETHBRIDGE on November 10, 2015



Moskvin et al.

23. Hall, E. J. The bystander effect. Health Physics 85, 31-35 (2003). DOI: 10.1097/00004032-200307000-00008 24. Baskar, R. Emerging role of radiation induced bystander effects: Cell communications and carcinogenesis. Genome Integrity 1, 13 (2010). DOI: 10.1186/2041-9414-1-13

25. van Waas, M., Neggers, S., Raat, H., van Rij, C. M., Pieters, R., van den Heuvel-Eibrink, M. M. Abdominal radiotherapy: a major determinant of metabolic syndrome in nephroblastoma and neuroblastoma survivors. PLoS One 7, 1-9, e52237 (2012). DOI: 10.1371/ journal.pone.0052237 Received: December 16, 2013; Revised: February 24, 2013; Accepted: February 28, 2014

Technology in Cancer Research & Treatment 2014 June 16. Epub ahead of print Downloaded from tct.sagepub.com at UNIV OF LETHBRIDGE on November 10, 2015

Effect of Scanning Beam for Superficial Dose in Proton Therapy.

Proton beam delivery technology is under development to minimize the scanning spot size for uniform dose to target, but it is also known that the supe...
4MB Sizes 1 Downloads 3 Views