Treatment planning system and beam data validation for the ZAP-X: A novel self-shielded stereotactic radiosurgery system.

Abstract

To evaluate the treatment planning system (TPS) performance of the ZAP-X stereotactic radiosurgery (SRS) system through nondosimetric, dosimetric, and end-to-end (E2E) tests. A comprehensive set of TPS commissioning and validation tests was developed using published guidelines. Nondosimetric validation tests included information transfer, computed tomography-magnetic resonance (CT-MR) image registration, structure/contouring, geometry, dose tools, and CT density. Dosimetric validation included comparisons between TPS and water tank/Solid Water measurements for various geometries and beam arrangements and end-to-end (E2E) tests. Patient-specific quality assurance was performed with an ion chamber in the Lucy phantom and with Gafchromic EBT3 film in the CyberKnife head phantom. RadCalc was used for independent verification of monitor units. Additional E2E tests were performed using the RPC Gamma Knife thermoluminescent dosimeter (TLD) phantom, MD Anderson SRS head phantom, and PseudoPatient gel phantom for independent absolute dose verification. CT-MR image registrations with known translational and rotational offsets were within tolerance (<0.5×maximum voxel dimension). Slice thickness and distance accuracy were within 0.1mm, and volume accuracy was within 0 to 0.11cm3 . Treatment planning system volume measurement uncertainty was within 0.1 to 0.4cm3 . Ion chamber point-dose measurements for a single beam in a water phantom agreed to TPS-calculated values within ±4% for collimator diameters 10 to 25mm, and ±6% for 7.5mm, for all measured depths (7, 50, 100, 150, and 200mm). In homogeneous Solid Water, point-dose measurements agreed to within ±4% for cones sizes 7.5 to 25mm. With 1-cm high/low density inserts, measurements were within ±4.2% for cone sizes 10 to 25mm. Film-based E2E using 4/5-mm cones resulted in a gamma passing rate (%GP) of 99.8% (2%/1.5mm). Point-dose measurements in a Lucy phantom with an ion chamber using 36 beams distributed along three noncoplanar arcs agreed to within ±4% for cone sizes 10 to 25mm. The RPC Gamma Knife TLD phantom yielded passing results with a measured-to-expected TLD dose ratio of 1.02. The MD Anderson SRS head phantom yielded passing results, with 4% TLD agreement and %GP of 95%/93% (5%/3mm) for coronal/sagittal film planes. The RTsafe gel phantom gave %GP of >95% (5%/2mm) for all four targets. For our first 58 patients, film-based patient-specific quality assurance has resulted in an average %GP of 98.7% (range, 94-100%) at 2%/2mm. Core ZAP-X features were found to be functional. On the basis of our results, point-dose and planar measurements were in agreement with TPS calculations using multiple phantoms and setup geometries, validating the ZAP-X TPS beam model for clinical use.