ESTRO 2024 - Abstract Book

S4025

Physics - Inter-fraction motion management and offline adaptive radiotherapy

ESTRO 2024

1 Aalborg University Hospital, Department of Medical Physics, Aalborg, Denmark. 2 Aalborg University, Department of Clinical Medicine, Aalborg, Denmark. 3 Aalborg University Hospital, Department of Oncology, Aalborg, Denmark

Purpose/Objective:

The manufacturing of bolus for superficial Radiotherapy can be challenging due to irregular surfaces. To achieve conformal shapes, bolus materials such as wax or, more recently, 3D printing can be used to create patient specific solutions. The material should prioritize patient comfort to reduce movement, and it must also be positioned consistently during treatment to ensure accuracy of the simulated dose distribution. In this study, we evaluated the bolus homogeneity and position consistency of wax and virtual planned 3D printed bolus, designed for Radiotherapy of the nose.

Material/Methods:

Twelve patients were referred for Radiotherapy, requiring the use of bolus to deliver the necessary dose to the skin. These patients were divided into two groups: One group (n= 6) was planned with wax bolus (Pinnacle dental wax) which was moulded before planning CT scan, while the other group (n=6) had bolus created within the treatment planning system for 3D printing (Formlabs 50A Resin). The bolus density, measured in Hounsfield Units (HU), was determined either from the planning CT scan (for wax bolus) or from an additional control CT scan, similar to the planning CT. Image guidance prior treatment sessions were achieved using Cone-Beam-CT (CBCT) with 6 degrees of freedom (6DOF) matching for bony structures. The volume of interest (VOI) encompassed the target volume and the anterior part of the cranial volume. The position of the bolus relative to the planned position was extracted through retrospective rematching exclusively for the bolus material. Deviations between treatment and planned positions were calculated as the difference between online and rematched position.

Results:

Comparable planning target volumes were observed between the group with wax and 3D printed bolus. The 3D printed bolus volumes were found to smaller than those of wax, with a mean reduction of 23.8 cc (95% CI: 8.9-38.9), determined through a two-sampled t-test with 5% significance level. Furthermore, the 3D printed bolus revealed greater homogeneity than wax in terms of reported HU variations (p < 0.01), as determined by a two-sample F-test with 5% significance level. The mean position accuracies for the two bolus materials were within 1mm and 1 degree, as shown in Table 1.

Table 1: Bolus mean deviations between treated and planned positions, divided in translations and rotations in terms of mean (sd) [middle 50%].