Is wax equivalent to tissue in electron conformal therapy planning? A Monte Carlo study of material approximation introduced dose difference

Ray R Zhang,Eduardo G Moros,Nikhil Rao,Vladimir Feygelman,Eleanor R Harris,Geoffrey G Zhang

doi:10.1120/jacmp.v14i1.3991

Abstract

With CT‐based Monte Carlo (MC) dose calculations, material composition is often assigned based on the standard Hounsfield unit ranges. This is known as the density threshold method. In bolus electron conformal therapy (BolusECT), the bolus material, machineable wax, would be assigned as soft tissue and the electron density is assumed equivalent to soft tissue based on its Hounsfield unit. This study investigates the dose errors introduced by this material assignment. BEAMnrc was used to simulate electron beams from a Trilogy accelerator. SPRRZnrc was used to calculate stopping power ratios (SPR) of tissue to wax, SPRwaxtissue, and tissue to water, SPRwatertissue, for 6, 9, 12, 15, and 18 MeV electron beams, of which 12 and 15 MeV beams are the most commonly used energies in BolusECT. DOSXYZnrc was applied in dose distribution calculations in a tissue phantom with either flat wax slabs of various thicknesses or a wedge‐shaped bolus on top. Dose distribution for two clinical cases, a chest wall and a head and neck, were compared with the bolus material treated as wax or tissue. The SPRwaxtissue values for 12 and 15 MeV beams are between 0.935 and 0.945, while the SPRwatertissue values are between 0.990 and 0.991. For a 12 MeV beam, the dose in tissue immediately under the bolus is overestimated by 2.5% for a 3 cm bolus thickness if the wax bolus is treated as tissue. For 15 MeV beams, the error is 1.4%. However, in both clinical cases the differences in the PTV DVH is negligible. Due to stopping power differences, dose differences of up to 2.5% are observed in MC simulations if the bolus material is misassigned as tissue in BolusECT dose calculations. However, for boluses thinner than 2 cm that are more likely encountered in practice, the error is within clinical tolerance.PACS numbers: 87.55.km, 87.56.ng

Highlights

Electron beams are commonly used for treatment of shallow tumors.[1]
It is often accomplished by the CT number threshold method.[11,12] Based on the CT number range, body regions are typically binned into air, lung, tissue, and bone. This is a reasonable approximation, as long as the dose is calculated to the media.[13]. The CT number of the machinable wax falls into the soft tissue range, and the bolus is treated as such despite the difference in chemical composition. This type of media misassignment can potentially lead to significant dose errors in Monte Carlo (MC) dose calculations.[11]. In this paper, we investigate how it affects BolusECT dosimetry
For the 12 MeV beam with a 3 cm wax slab, the dose in tissue just under wax is overestimated by 2.5% ± 0.7% (1 SD) if wax is treated as tissue

Summary

Introduction

Electron beams are commonly used for treatment of shallow tumors.[1]. To compensate for body surface irregularities and/or to modulate the therapeutic depth, a bolus is often employed.[2,3,4] The chest wall[4,5] and head and neck[6] are among the sites treated with this technique.Required bolus shape depends on beam energy, external body contour, and target volume. Assuming similar linear electron stopping power between the bolus material and tissue and ignoring multiple Coulomb scattering, a simple approach makes the thickness of the bolus along a fan line such that the depth from the surface to the distal edge of the target volume remains constant.[5,7] More recent computer-aided bolus design uses a three-dimensional (3D). As a part of the quality assurance (QA) process, a new CT scan with the bolus in place is acquired, and the dose is recalculated. It should closely match the original plan where the bolus is represented by a region of interest assigned the bolus material density.

Methods

Results

Conclusion