Author: Maria de la Luz De Ornelas, Yunze Yang 👨🔬
Affiliation: University of Miami 🌍
Purpose:
An extended Winston-Lutz test (EWL), also known as off-axis Winston-Lutz, was performed to assess the linear accelerator’s (linac) deviations at isocenter and five off-iso targets. The impact of imaging setup on test results was evaluated to improve understanding of the linac’s capabilities for single-isocenter and multiple-target delivery.
Methods:
The deviations from imaging and radiation isocenter were verified using MV and kV imagers, confirming that the isocenter alignment was within clinically acceptable tolerance. The SNC MultiMet-WL phantom and software were used for off-iso WL tests. We setup the phantom utilizing either MV-MV image pairs or CBCT imaging. MV images aligned phantom to the machine’s radiation isocenter (rad-iso), while CBCT aligned it to the imaging isocenter (img-iso). EWL was conducted three times for each setup (rad-iso and img-iso). Deviations from expected positions were calculated using MultiMet software v2.1. Targets were positioned as follows (IEC61217): Target 1 (+3 cm y), Target 2 (+1.5 cm y, +3 cm x), Target 3 (-3 cm x), Target 4 (-5 cm y, -3 cm x), and Target 5 (-7 cm y). A paired student t-test was used for statistical comparison.
Results:
The maximum deviation for the MV and kV imagers were 0.024 cm and 0.025 cm, respectively, at the isocenter. Img-iso setups showed larger deviations than rad-iso setups for off-iso targets, with significant differences for Targets 4 and 5 (p=0.0167, and p=0.0004 respectively). Target 5 exceeded the 1 mm threshold in both setups, and Targets 3 and 4 exceeded tolerances when aligned using img-iso.
Conclusion:
Single-isocenter, multi-target treatments require awareness of linac limitations, including differences between radiation and imaging isocenters. Img-iso setups can lead to greater deviations for off-isocenter targets. Our results suggested that target volume expansion may be necessary to mitigate these effects on distant off-isocenter targets.