Investigation of effect of filter on the stand‐up technique for total skin irradiation by Monte Carlo simulation

Abstract Purpose The aim of this study was to investigate dosimetric effects of scattering filter on the stand‐up technique for total skin irradiation (TSI) with a single electron field by Monte Carlo (MC) simulation. Methods MC simulations were performed with BEAMnrc and DOSXYZnrc packages under EGSnrc environment. Scattering filter of a metal disc was mounted in the accessory slot. The filter materials (Cu, Fe, Au, Zn, Ag) were investigated, with thickness ranging from 0.05 to 0.55 mm, depending on material. The extended source to skin distance (SSD) ranging from 250 to 350 cm was studied. The following dosimetric quantities were evaluated: percent depth dose (PDD), profiles and output factor at depth of maximum, and composite dose distribution on a 30‐cm diameter cylindrical phantom. They were compared with the standard dual beam technique used at our clinic. The effects on different patient sizes were also studied. Results No filter produced acceptable profile flatness (±10% within the central 160 cm) at 250 cm SSD. At 300 cm SSD, Au (0.1 mm), Ag (0.25 mm), Cu (0.5 mm) produced acceptable flatness while Zn (0.45 mm) required 325 cm SSD. For these four configurations, the dmax was 0.90–0.99 cm, similar to dual beam (0.97 cm); R50 was 1.85–1.91 cm, compared with dual beam of 2.06 cm; the output factor ranged from 0.025 to 0.029, lower than the dual beam (0.080). With the composite fields for four configurations, the dmax was 0.10 cm, compared with dual beam (0.16 cm). The surface dose was 97%, similar to dual beam (96%). B‐factor was 3.3–3.4, compared with dual beam of 3.1. The maximum X‐ray contamination was 3%, higher than dual beam (1%). Conclusions The investigation suggests the TSI stand‐up technique can be implemented using a single electron beam if a customized filter is used. More dosimetric measurements are needed to validate the MC results and clinical implementation.


| INTRODUCTION
Total skin Electron irradiation (TSEI or TSI) is an external beam therapy used to treat patients with malignant skin diseases such as mycosis fungoides or cutaneous T-cell lymphoma. It is a special electron therapy technique that involves delivering a homogeneous radiation dose to the entire skin over the whole body within a limited depth (few millimeters), 1 while sparing the radiation dose delivered to the organ at risks (OARs) beyond a few centimeters depth. To deliver a successful total skin electron therapy, the American Association of Physicists in Medicine Task Group No. 30 (AAPM TG-30) 2 recommends that the field size of the composite electron field must be approximately 200 cm in height by 80 cm in width at the treatment plane to cover large patients, the dose uniformity over the central 160 × 60 cm 2 region should be within ±8% in vertical and ±4% in horizontal directions, and X-ray contamination (1%) of the electron fields is desired. 2,3 Various TSI techniques have been developed and described in AAPM TG-30. 2 At our clinic, the standard procedure for TSI treatment is the Stanford six dual-field method. 2 To provide a uniform dose distribution on the patient, dual electron fields with ±19 degree angled from horizontal are directed at patient standing on the TSI platform at an extended source to skin distance (SSD) of 300 cm. Clinic. 4 The lay-down technique was first developed by Wu et al. 5 and further modified by Deufel  One key requirement in the lay-down technique is the use of a customized scattering filter to broaden the electron field for compensating the reduced SSD for the vertex fields. Few research groups have implemented different designs of scattering filter for TSI techniques: Pham et al. 3 designed an aluminum/polystyrene electron scattering filter for a single field rotational total skin irradiation to redistribute the electron beam; Podgorsak et al. 7 constructed an electron beam degrader made of lead/aluminum; El-Khatib et al. 8 designed a beam scattering filter from Lucite; Reynard et al. 9 built a custom flattening filter constructed of aluminum, lead, and polymethyl methacrylate used in rotational total skin electron irradiation.
According to the previous studies, it could be more efficient to produce an equivalent beam uniformity by a scattering filter with a complicated design and variations of material and thickness.

| MATERIALS AND METHODS
In this section, the design of the scattering filter and the description of the modified TSI stand-up are first presented. The MC model implemented under EGSnrc environment to investigate the dosimetric effect of the scattering filter is then described in detail. Furthermore, dosimetric metrics used to analyze and compare the standard and modified TSI techniques are discussed. The experimental measurement for the modified technique is also described in the end of this section.

2.A.1 | Scattering filter design
The customized scattering filter used in this study was based on the design of the scattering filter originally from Mayo Clinic, 4 constructed by a 0.25-mm thickness copper disc placed between two 1 mm polycarbonate layers and implemented for a TSI lay-down technique at our clinic. The detailed geometry of the scattering filter is shown in Fig. 1(a).
To find out the optimal configuration of scattering filter, various filter material, filter thickness, and setup SSD were studied (Table 1) The metals, iron (Fe), gold (Au), silver (Ag), zinc (Zn), and copper (Cu) were chosen since they are high atomic number material which are readily available on the market and physically and chemically stable at room temperature. The filter thickness from 0.05 to 0.55 mm and the setup SSD from 250 to 350 cm were studied. For each material with different thickness, the profile flatness at each extended SSD was examined to evaluate if it can achieve the requirement recommended by AAPM TG-30. 2

2.A.2 | Modified TSI stand-up technique
The customized scattering filter was mounted onto the accessory slot of the Linac (57.5 cm from the beam target). Instead of six pairs of electron fields (upper and lower fields at each direction) that were used in the standard technique, six single electron fields were directed to phantom standing at an extended SSD in the modified standup technique. Phantom was rotated along the cranial-caudal axis on a TSI platform in six positions with 60°interval to get full dose coverage to skin. For each direction, a single electron field with jaws opening (X/Y Jaws) at 30 × 40 cm 2 , collimator rotation at 90°, and fully retracted multileaf collimator (MLC) were used for the modified stand-up technique.

2.B | Monte Carlo simulation
In our previous study, 6 we built a MC model that has been successfully used to validate both TSI stand-up and lay-down techniques.   Fig. 1(b). In our study, the Linac geometry for the 6 MeV electron beam was composed of secondary collimators (X and Y Jaws) and the scattering filter which were modeled by JAWs and | 139 SLAB/FLATFILT CMs, respectively. A CM of SLAB was specified for the position of phase-space file which served as a main beam source.
A scoring plane, where any particle traveling through it will be recorded and stored in an output phase-space file, can be defined and placed under any pre-built CMs.
To simulate the modified stand-up technique, scoring planes were defined in BEAMnrc and placed at 58 cm (Linac exit window), 250, 300, 325, and 350 cm from the beam target. phase-space files generated at the Linac exit window was used to analyze the beam characteristic using BEAMDP code (BEAM Data Processor). 15 The

2.C | Comparison metrics
To evaluate the effect of the scattering filter, dosimetric metrics such as beam profiles at depth of maximum dose, d max , percentage depth dose (PDD), and output factor were computed and compared with the simulated data for the standard TSI stand-up technique that has been used at our clinic. The study of these dosimetric metrics were conducted by EGSnrc MC simulations, with some actual measurements in phantom to validate MC results (Section 2.D).

2.C.1 | Phase-space file analysis
Mean energy distribution and angular distribution of phase-space file generated at the Linac exit window (57.5 cm from beam target) were analyzed to evaluate the effect of an existence of the scattering filter on beam characteristics using BEAMDP.     Table 3.   However, the scattering filter can also introduce secondary photons which will produce a higher photon contamination in the patient. The overall effect depends on the filter and the beam arrangement. The modified technique produced a maximum X-ray contamination of 3% due to the effect of the scattering filter on beam characteristic (shown in Fig. 2), slightly higher than the standard technique (1%) with dual open fields, which was intentionally angled away to minimize the X-ray contamination on the central axis. In our lay-down technique with 0.25 mm Cu filter, the maximum X-ray contamination was about 2%. Because only single field per direction was used in the modified technique, the X-ray contamination was highest at the central slice, but it quickly decreased with the off-axis distance. For example, at the off-axis distance of 10 cm, the X-ray contamination dropped to 2%, as shown in Fig. 6. Nevertheless, a further evaluation of X-ray contamination on the modified technique should be investigated to find out if a higher X-ray contamination would result in any extra side effects on patients.

4.B | Dose discrepancy on the beam profile between the measurement and MC result
As shown in Fig. 3(a), there was a fairly large difference in the longitudinal dose profile between the measurement and MC results for the 0.5 mm Cu filtered field, up to 10% at 60 cm away from central axis.
In our previous study, the MC results for the lay-down technique which also implemented a scattering filter in the MC model in general agreed well with the measurement. Therefore, the dose discrepancy on the profile might be caused by the research scattering filter (0.5 mm Cu) and the mounting plates used in the measurement. The 0.5 mm Cu disc from Amazon.com might not be suitable for this study, even though it was rated at 99.9 % purity. That is, its purity and exact geometry might not be as good as the scattering filter clinically used for the lay-down technique. Therefore, more accurate constructions in terms of filter thickness, mounting plates, and pure material from a professional manufacturer and rigorous measurements are needed if the modified technique is clinically implemented.
The uncertainty for the MC simulation included room scatter effects 16 from walls, floor, ceiling, and TSI platform in the treatment room since the MC model cannot account for every detailed geometry. Furthermore, the measurements done at the treatment plane can be hard to set up accurately. That is, any variations of SSD, angular position, and off-axis distance at an extended SSD could bring uncertainty to the measurement. Therefore, the uncertainties from both MC simulation and measurement mentioned above could potentially lead to the discrepancy shown on the dose profiles.

| CONCLUSIONS
The results of our Monte Carlo calculations found that the TSI stand-up technique can be implemented using a single electron field if a customized filter is used, with producing comparable dosimetric results as the standard technique that has been used at our clinic.
Monte Carlo simulation is valuable in performing this type of investigation, it reduces the need of measurement considerably. In addition to those measurable quantities, the Monte Carlo simulation can provide further investigations such as a full dose distribution of the patient phantom, and the ability to investigate and optimize techniques such as different filter designs, SSD, and field size variations.

ACKNOWLEDG MENT
The authors thank Tim Eaton for the assistance in the construction of the filter used in the experiment.

CONFLI CT OF INTEREST
There is no conflict of interest for all authors.