ESTRO 2025 - Abstract Book

S3105

Physics - Inter-fraction motion management and offline adaptive radiotherapy

ESTRO 2025

1419

Digital Poster Open or closed: a prospective, self-controlled study of the performance of two immobilisation mask types in head and neck cancer radiotherapy Erik Lundin 1,2 , Ulf Granlund 3 , Per Thunberg 3 , Johan Ahlgren 2 1 Oncology department, Örebro University hospital, Örebro, Sweden. 2 Faculty of Medicine and Health, Örebro University, Örebro, Sweden. 3 Department of Medical Physics, Örebro University hospital, Örebro, Sweden Purpose/Objective: This study aimed to evaluate the effectiveness of open versus closed immobilisation masks for initial patient positioning and for maintaining position during radiation treatment for head and neck cancer. There is a limited amount of research published on this topic.[1-3] A related investigation into patient experiences with these mask types using the same study sample has been published previously.[4] Material/Methods: The study included 20 patients scheduled for 34 fractions of head and neck cancer radiotherapy, using masks to immobilise the head, neck and shoulders. (Figure 1) Each patient was fitted with both an open and a closed mask, alternating between them weekly, with the first mask type assigned randomly.

For the open mask, initial positioning utilized a surface scanning system followed by orthogonal kilovolt (kV) imaging. For the closed mask, positioning was guided by mask markings aligned with in-room laser lines, followed by orthogonal kV imaging. Post-treatment kV images were also acquired to measure movements during the treatment. All positioning images, taken before and after treatment, were manually matched to a reference image by an experienced specialist nurse. The reference image was derived from the dose-planning CT. Movement parameters were recorded in the vertical, longitudinal, and lateral directions. Due to a change in clinical imaging protocol during the study, 7% of setup images were not matched as initially planned, and only one-third of post-treatment images were matched. The presented analyses use only data matched per protocol.

Results: The need for position correction after initial setup, measured with kV image matching, was analysed for 278 open