Room: Karl Dean Ballroom A1
Purpose: One potential risk for single-isocenter multiple-target VMAT SRS is the increased geometric misalignment between the actual treatment field and planned field for off-isocenter targets. This phenomenon of field shift is mainly caused by mechanical uncertainty and asymmetric penumbra. As the dose gradient is about 13~18%/mm for SRS plans, a 1-mm geometric misalignment will result in a significant dose deviation. Here, we present a systematic methodology to quantify the field misalignment and assess its impact on CTV margin definition. A literature search indicates that this is the first such systematic evaluation on a TrueBeam machine.
Methods: Fourteen test plans were computed with EclipseV13. Each test plan contained MLC-defined static fields (3x3cm) with different combinations of gantry, collimator, couch angles, and off-isocenter distances (0, 2, 3, 4, 5, and 7 cm). A total number of 448 fields were created. All fields were measured with MV imaging on a TrueBeam machine. A Python program was written to batch process the images.
Results: We found that field center misalignment changed with off-isocenter distance. It was primarily the maximum misalignment that increased with off-isocenter distance, while the mean and minimum misalignments remained fairly constant. Few fields had large field center shifts, while some even had decreasing center shifts as off-isocenter distance increased. We also found that mechanical rotations did not induce any systematic field misalignment when off-isocenter distance ≤ 3 cm. The field misalignment became apparent when off-isocenter distance was ≥ 5 cm. Finally, we determined that the ratio of measured/planned field size was primarily affected by MLC leakage and not by off-isocenter distance.
Conclusion: Based on the results of this study, we recommend an additional 0.4 mm margin to CTV when off-isocenter distance is ≤ 3 cm and 0.6 mm margin to CTV when off-isocenter distance is ≥ 4 cm for TrueBeam machines.