Evaluation of the dosimetric effect of scattered protons in clinical practice in passive scattering proton therapy

Abstract The present study verified and evaluated the dosimetric effects of protons scattered from a snout and an aperture in clinical practice, when a range compensator was included. The dose distribution calculated by a treatment planning system (TPS) was compared with the measured dose distribution and the dose distribution calculated by Monte Carlo simulation at several depths. The difference between the measured and calculated results was analyzed using Monte Carlo simulation with filtration of scattering in the snout and aperture. The dependence of the effects of scattered protons on snout size, beam range, and minimum thickness of the range compensator was also investigated using the Monte Carlo simulation. The simulated and measured results showed that the additional dose compared with the results calculated by the TPS at shallow depths was mainly due to protons scattered by the snout and aperture. This additional dose was filtered by the structure of the range compensator so that it was observed under the thin region of the range compensator. The maximum difference was measured at a depth of 16 mm (8.25%), with the difference decreasing with depth. Analysis of protons contributing to the additional dose showed that the contribution of protons scattered from the snout was greater than that of protons scattered from the aperture when a narrow snout was used. In the Monte Carlo simulation, this effect of scattered protons was reduced when wider snouts and longer‐range proton beams were used. This effect was also reduced when thicker range compensator bases were used, even with a narrow snout. This study verified the effect of scattered protons even when a range compensator was included and emphasized the importance of snout‐scattered protons when a narrow snout is used for small fields. It indicated that this additional dose can be reduced by wider snouts, longer range proton beams, and thicker range compensator bases. These results provide a better understanding of the additional dose from scattered protons in clinical practice.


| INTRODUCTION
Charged particles, including protons and carbon ions, are being increasingly introduced into modern radiation therapy. The physical and biological characteristics of proton beams, including a low entrance dose and the absence of an exit dose, make proton beams more attractive than conventional photon beams in radiation therapy. 1 At present, more than 80 proton therapy facilities are in operation worldwide, with the number increasing every year. 2 Despite the clinical use of proton therapy for many years, several uncertainties remain. Conventional treatment planning systems (TPS) have limitations in dose calculation for proton therapy. Uncertainties in dose calculation may be caused by, for example, conversion of Hounsfield units (HU) to relative proton stopping power, limited modeling of multiple Coulomb scattering, biological effects, and scattered and secondary radiation from the treatment nozzle. [3][4][5][6] Due to the inherent limitations of conventional pencil-beam algorithms, scattered protons from a field-limiting aperture or collimator have not been fully included in dose calculation by commercial TPSs for passive scattering proton therapy. 7,8 Previous studies focused on the dosimetric influence of the scattered protons from the aperture, which is mounted onto the end of the nozzle to shape the proton beam onto the target, because it is likely the most significant contributor to proton scattering.
The influence of scattered protons from slit collimators in small proton fields between 2 mm and 20 mm was investigated by Monte Carlo simulation for a proton beam of 150 MeV. 9 These simulations showed that the contribution of collimator scattered protons was not negligible, constituting 20% of the total dose at the exit of the collimator and 5% of the total dose even 15 cm away from the collimator. This influence of scattered protons from the edge of the collimator aperture was also detected in a series of studies of lowenergy proton beams used in ocular proton therapy. [10][11][12] The size of the final collimator aperture was a little larger (24 mm) than in the previous study, and the measured and simulated depth-dose profiles showed the contribution of scattered protons at shallow depths. Following these studies, emphasizing the dosimetric influence in small proton fields, Titt et al 13 investigated that the dependency of the dosimetric impact of collimator-scattered protons on several variables, including proton beam range, the modulation width of the spread-out Bragg peak (SOBP), field size, and the air-gap between the collimator and the phantom.
Investigations of the effects of aperture-scattered protons led to the development of extended algorithms that included this scattering from the aperture to more precisely calculate patient dose. 12,[14][15][16] In addition, new aperture structures that produce fewer scattered protons were developed. 17,18 But despite these attention to aperturescattered protons, they are not considered in commercial TPSs. The dosimetric effects in clinical practice, when the range compensator is included, have not been reported. 13 The present study was designed to evaluate the potential dosimetric effects of scattered protons from field-limiting apertures in clinical practice, including the range compensator in passive scattering proton therapy. By comparing the measured dose with doses calculated by TPS and Monte Carlo simulation, the effect of the scattered protons could be determined, even after these protons passed through the range compensator. The filtration of this effect by the structure of the range compensator was investigated, and the contribution of a snout and an aperture to the dose of scattered protons was analyzed by Monte Caro simulation. In addition, the effects of snout size, beam range, and thickness of the range compensator base on the dosimetric effect of scattered protons were evaluated.

2.A | Monte Carlo simulation of the proton treatment nozzle
A Monte Carlo simulation system of a proton therapy nozzle in double-scattering delivery mode at the National Cancer Center Korea ( Fig. 1) was developed for independent dose verification of treatment plans. 19 The design of the model nozzle was based on the manufacturer's blueprints (the IBA universal nozzle, IBA) and a previously designed model based on Geant4. 20,21 The tool for particle simulation (TOPAS, 3.1.p03 version) 22 was used for nozzle modeling and simulation of proton transport. In the previous study, 19 the developed Monte Carlo simulation model of the scattering proton therapy nozzle was validated by comparison of its simulated percent depth-dose (PDD) profiles with previously measured PDD profiles.
In addition, the results calculated by the Monte Carlo simulation and TPS for a lung phantom, an internal mammary node, and an abdomen were compared to verify the feasibility of the developed system in clinical practice. 19 The developed Monte Carlo model showed good agreement with both the actual measurement and TPS calculation, differing less than 1 mm in range and modulation width. 19 In this study, the dose distribution in a uniform water phantom was calculated using the Monte Carlo simulation system. Virtual treatment plans and their beam parameters, which were determined by the converting algorithm (Convalgo, IBA), were imported into the simulation system to set up the beam nozzle. In the simulation, protons were assumed to be emitted from the beam entrance with the determined mono-energy and passed through the preset nozzle structures. The energy, positional, and angular spread of the initial proton beam were assumed as a single Gaussian distribution with parameters benchmarked by measurements performed in the previous study. 21 The apertures and range compensators in the simulation were modeled on the designs included in the imported DICOM files and were composed of brass and poly(methyl methacrylate) (PMMA) ( Table 1). 19 The phantom was placed at the same distance from the range compensators as the air gaps in the virtual plans. The material of the phantom was assigned as G4_Water. The other simulation parameters were the same as those in the reference. 19 The simulation parameters are briefly listed in Table 2.
To ensure sufficient statistical accuracy, each simulation included more than 10 9 primary protons. The secondary neutrons were not

2.B | Treatment plans for validation and evaluation
Virtual treatment plans were used to evaluate and analyze the dosimetric influence of scattered protons from the nozzle. These plans    Table 3) and was normalized to deliver 200 cGy to the center of the PTV in a single fraction. The other parameters, including the beamline properties (AP direction, range 7.31 cm, and SOBP width of 6.21 cm) and air gap (8.5 cm), were the same as those in the original plan (Proximal 100). The aperture and range compensator were identical to those in the original plan.
To assess the effect of scattered protons with depth, the 2dimensional dose distribution in the solid water phantom was measured at several depths using the IBA proton therapy machine installed in the second proton treatment room at the National

Material
Density (g/cm 3     The discrepancy at the center was noticeable, but was not observed in the gamma analysis between the measured and simulation results. These unexpected additional doses in the central region at shallow depths were likely caused by protons scattered from the snout or aperture. Figure 5 shows a comparison of the calculated PDD profiles on TPS with the profiles from Monte Carlo simulation. All profiles were acquired on the CAX and normalized to the dose at  showed that the contribution of snout-scattered protons to the additional dose (4.21%) was almost twice as high as the contribution of aperture-scattered protons (2.1%). This was reasonable, as the snout (60 cm) was much longer than the aperture (6.5 cm). Taken together, these findings indicate that the snout, not the aperture, is the major source of scattered protons in this plan.  Monte Carlo simulations, and measurements (Fig. 3, 4).

3.C | Effects of snout sizes and beam ranges
The effects of snout sizes and beam ranges on proton scattering were investigated. Figure 9 shows Moreover, the additional dose at shallow depths was found even when the proton beam range increased. Figure 10 shows the PDD profiles on CAX in TPS calculations and Monte Carlo simulations for virtual treatment plans with the same snout (SNT 100) but proton beams of different ranges (Table 3) Because these plans were established just for comparison of the F I G . 6. Simplified diagram of dose accumulation by (a) primary protons and (b) protons scattered from the snout or aperture, and considering the range compensator in the virtual treatment plan.
dose in the proximal regions, the plan parameters were not optimized for each plan. It seems that enough smearing margin is required to the range compensator for covering the distal targets.

3.D | Effects of the compensator base on the dosimetric effects of scattered protons
The effects of the compensator based on the effects of scattered protons were also analyzed in measurements and Monte Carlo simulations. Virtual treatment plans covering the same target volume in the Proximal 100 plan, but with different compensator bases, were used ( Table 4). The resulting IDD profiles normalized to the dose at the center of the SOBP (a depth of 4.2 mm) were compared.

| DISCUSSION
The dosimetric effect of protons scattered from the snout and aperture passive scattering proton therapy was measured and evaluated in clinical practice when a range compensator was included. The results of this study confirmed previous results based on Monte Carlo simulations, 9,13 and showed that the effect of scattered protons can persist even after passing through a range compensator.
These analyses indicated that the dosimetric effect of scattered protons is more complicated than previously thought, and requires more detailed consideration in practice.
The snout used to insert the aperture and the range compensator has another role to protect patients from unwanted radiation. The structure and material composition of the snout vary by supplier and site, but it is usually composed of high-Z materials. The snouts can make a non-negligible contribution to proton scattering, depending on their design. The combination of a large aperture opening and a narrow snout can be considered an extended aperture. Increases in aperture length have been associated with increases in the dose from scattered protons. 13 In this study, the use of a long and narrow snout,

D A T A A V A I L A B I L I T Y S T A T E M E N T
The data that support the findings of this study are openly available in [Default Parameters] at [https://topas.readthedocs.io/en/latest/], reference number [23].