SU‐E‐T‐416: Targeting Accuracy of a TrueBeam System for Multi‐Lesion Cranial Radiosurgery with Single Isocenter

Purpose: To investigate the targeting accuracy of a Varian TrueBeam System for delivering intracranial radiosurgery to multiple lesions using a single isocenter (ISO). Methods: The study was performed using hidden target (BB) tests along with a MV portal imager (EPID). Six BBs were inserted into a head phantom simulating the locations of multiple metastatic tumors. One BB acted as the common ISO. The distances from the ISO to the other BBs were 2.9 to 7.1 cm. Simulating the actual patient treatment in our clinic, the phantom was immobilized using a BrainLAB U‐shape frame and a head mask. CT image was acquired with 1 mm slice thickness. For a selected couch, gantry and collimator angle combination, a 3×3 cm MLC‐shaped field was created for each BB. The centers of BB and the field were aligned, and the distance from the plan was read out on the DRR. Before shooting the tests, CBCT was used to setup the phantom, and the translational/rotational mismatch was corrected iteratively until less than 1 mm and 0.1 degree respectively. The actual positions of the BBs relative to their corresponding field centers were measured on the portal images. The deviation between the measured and planned BB positions quantifies the geometric targeting error. The test was performed for various couch, gantry and collimator angle combinations (total 88 tests). Results: The mean±SD of the BB position deviations were 0.0±0.4 and 0.2±0.4 mm in the image transverse‐and longitudinal‐direction respectively; and 0.5±0.3 mm for the vector length. Six of 88 BB tests (6.8%) exceeded the vector length of 1 mm with a maximum of 1.2 mm Conclusion: The mechanical targeting accuracy of the TrueBeam System at our institution is acceptable for delivery of intracranial radiosurgery with a single isocenter for treatment of multiple targets. Application of this technique is being investigated.