Session Item

Immobilization with an open face mask is more comfortable and less invasive than frame-based immobilization during radiosurgery, but concerns about intra-fraction motion must be addressed, particularly when treating multiple brain metastases with a single isocenter. The online setup control with surface gating has been shown to improve the safety of SRS treatments. We developed an institutional protocol, evaluated the accuracy of the patient setup, and analyzed the impact of setup uncertainties on the dose distribution, PTV coverage and OAR sparing.

Fifteen patients treated in a single fraction and single isocenter for multiple brain metastases (2 to 8 lesions) were included in this study. The VMAT-plans were calculated with Varian HyperArc and delivered on a TrueBeam clinac equipped with an optical surface monitoring system (OSMS, Align-RT, Vision-RT). The patients were fixed with an open face mask on a dedicated board (QFix Encompass), the same immobilization setup was used for MR-imaging directly after the CT-simulation. A GTV to PTV margin of 2 mm was used for contouring. The prescribed dose was 18 to 20 Gy at the encompassing 80%-isodose.
The patients were initially set up according to the displacement shown by the OSMS, the positioning was then adjusted by IGRT. Patient positioning was controlled online, using as baseline a reference surface acquired immediately after the setup-CBCT. If the real time shifts detected be the OSMS exceeded 1mm/1°, the beam was automatically stopped and IGRT-repositioning was performed. The real time shifts were recorded during the entire treatment. Finally, an end-CBCT was acquired. The planning-CT was shifted and rotated according to the displacements detected by the end-CBCT and the HyperArc plan was recalculated on the displaced planning-CT.

The displacement shown by the OSMS during the initial patient positioning correlated very well to the setup CBCT shifts (translations 0.6 ± 0.4 mm, rotations 0.5 ± 0.3°). The OSMS-detected real time shifts did not show a significant dependence on the couch rotation, with a trend towards bigger shifts for couch angle 90°/270°. The average shifts between the setup-CBCT and the end-CBCT were 0.3 mm / 0.3° (with a maximum shift of 1.1 mm / 1.2° for one patient). The recalculated plans showed for all patients a GTV coverage ≥ 99.5%. The average PTV volume encompassed by the prescribed isodose was 94% (worst coverage 90% for the outlier patient). No relevant difference in the OAR-dose was observed.

Surface gating allows a safe delivery of HyperArc treatments of multiple brain metastases with a single isocenter, if the tolerances for the online monitoring are sufficiently small. The clinical outcome will be assessed in a longer follow-up.