ESTRO 2024 - Abstract Book

S5936

RTT - Treatment planning, OAR and target definitions

ESTRO 2024

110

Digital Poster

Affect of auto feathering and junction overlapping length with plan robustness of CSI

JINHUEI JI 1 , Ting-Chun Lin 1 , Shu-Sheng Li 1 , Ji-An Liang 1,2 , An-Cheng Shiau 1,3 , Ti-Hao Wang 1

1 China Medical University Hospital, Department of Radiation Oncology, Taichung, Taiwan. 2 China Medical University, Department of Medicine, Taichung, Taiwan. 3 China Medical University, Department of Biomedical Imaging and Radiological Science, Taichung, Taiwan

Purpose/Objective:

Craniospinal irradiation (CSI) is a sophisticated radiotherapy planning technique. Incorrect field matching can lead to dosimetric irregularities and uncertainties. Varian Eclipse has introduced an auto feathering function for extended beam radiotherapy to enhance dose uniformity across junctions. However, the effectiveness of this function is influenced by the field overlap area. This study aims to ascertain the optimal fields overlap area in clinical CSI conditions.

Material/Methods:

We retrospectively collected nine case and re-plan Volumetric modulated arc therapy (VMAT) CSI plans, each incorporating three isocenters placed in cranial, upper-spinal, and lower-spinal regions. The auto feathering algorithm was applied with on and off settings, and field overlap areas of 3, 4, 5, 7, 9, and 10 cm were tested. Patients were subjected to up to 5 mm shift in the cranial-caudal direction to simulate setup uncertainties. Two field overlap regions of the planning target volume (PTV) were taken for plan robustness evaluation.[WU1] Dose distribution was assessed using metrics such as V95%, D2%, D98%, organ at risk (OAR) doses, conformity index (CI), and homogeneity index (HI).

Results:

In the absence of a center shift, no significant differences in planned target volume dose, CI, HI, and organ at risk doses were observed between auto feathering on and off mode. During the robustness study, auto feathering significantly reduced hot spots (D2%) and enhanced target coverage (D98%) compared to the absence of auto feathering. Implementation of auto feathering with larger overlapping lengths improved plan robustness.[WU1] Shifts of 5 mm in the cranial-caudal direction resulted in < 3% dose deviations when field overlap exceeded 5 cm and auto feathering was active.

Conclusion:

The utilization of the auto feathering function effectively mitigates dose distribution changes induced by setup uncertainties. This study demonstrates substantial improvements in plan robustness, particularly with extended field