KEY POINTS
- The retrospective analysis included 669 fractions from 20 patients receiving curative head and neck VMAT to 66–68 Gy in 33–34 fractions using thermoplastic mask immobilization.
- Initial positioning used CBCT with six-degree-of-freedom correction, followed by ExacTrac Dynamic X-ray verification. During treatment, SGRT continuously monitored the surface and triggered beam hold and X-ray verification for deviations exceeding 2 mm.
- Residual ExacTrac corrections after CBCT were small on average: −0.2 mm vertical, +0.8 mm longitudinal, and −0.3 mm lateral.
- CBCT roll was strongly associated with subsequent lateral correction, with a slope of −0.98 mm per degree, while pitch predicted longitudinal correction at +0.49 mm per degree. A 3° rotational difference could therefore correspond to translational offsets approaching 3 mm.
- SGRT triggered additional image guidance during 100 of 669 sessions (15%). Events were distributed relatively evenly across the three treatment arcs.
- Only 52% of SGRT-triggered events were confirmed on X-ray imaging to exceed the 2 mm threshold. Nearly half were therefore false-positive surface alerts or surface changes not representing clinically relevant internal displacement.
- The maximum observed intrafraction displacement was 8 mm, supporting the potential value of continuous monitoring despite imperfect specificity.
CLINICAL TAKEAWAY
Continuous SGRT adds useful surveillance between CBCT acquisitions and can catch meaningful head and neck movement during beam delivery. However, approximately half of alerts would have led to unnecessary intervention without X-ray confirmation, so surface monitoring and radiographic IGRT should be used as complementary tools.
SOURCE
Technical Innovations & Patient Support in Radiation Oncology