Comparison of respiratory motion management by means of three abdominal corsets for particle therapy

Purpose/Objective:

Particle therapy (PT) has the potential of improving the outcome in radiotherapy (RT) due to its inverse dose profile and the superior sparing of healthy tissues surrounding the target volume compared to photon therapy. However, PT is strongly susceptible to anatomical changes, and especially for the treatment of abdominal tumours, strategies for motion management are required.
The purpose of this study was to investigate and compare the potential usability of three different abdominal corsets in PT by measuring their water equivalent ratio (WER) in proton therapy as well as by analysing their effect on the respiration-induced motion of the pancreas.
Material/Methods:
The corsets differed in terms of geometry (thickness of 2.5mm – 24mm), material (polyethylene (PE) vs. polyurethane (PU)) as well as regarding the method of construction (patient individual vs. patient independent). A healthy volunteer was scanned on a 1.5T MR scanner (Magnetom Aera, Siemens Healthineers) on two consecutive days while he was wearing each of the three respective corsets and without him wearing a corset for reference. A gradient echo sequences with radial golden angel acquisition was used and reconstructed to a 4D data set with 20 phases. The pancreas was delineated in max exhale and max inhale phase using the open-source software MITK (Fig. 1). The centre of mass was calculated as a surrogate for the respiratory motion of the pancreas in each of the four scenarios for both days.
After acquiring CT scans of the three corsets for assessment of material homogeneity and regularity of material thickness, WER measurements were performed at two different proton energies (150MeV, 200MeV) using a multi-layer ionization chamber (Giraffe, IBA Dosimetry) to measure the shift of the single beam Bragg Peak after penetrating the corset sample.
Results:
All three abdominal corsets led to reduced pancreatic motion, and the effect was largest in inferior-superior direction (Table 1). The CT revealed a perfectly homogeneous material for the two PE corsets with a constant thickness of 2.5±0.1mm and 4.9±0.1mm. In case of the PU corset the material was inhomogeneous with air inclusions throughout the whole corset. Furthermore, its thickness varied between 8.0-24.2mm in the relevant region. The WER of the two polyethylene corsets was determined to be 0.990 and 0.956, while the WER of the polyurethane corset was 0.298.
Conclusion:
While all three corsets reduce the respiratory motion to a similar amount, the material analysis revealed that the polyurethane corset is not suitable for PT due to its inhomogeneous structure and irregular thickness. On the other hand, the two PE corsets both show very stable material conditions which could, in terms of physics, easily be included in treatment planning and a fractionated treatment scheme. However, due to their different construction approach, the PE corsets have respective benefits in accuracy of fit, flexibility, cost and the time required for preparation.