The InCise™ MLC for CyberKnife showed high accuracy and adequate characteristics for SRS/SBRT applications. MLC performance after exercise demands specific quality assurance measures.

Purpose: To assess characteristics and performance of the “Incise™” MLC (41 leaf pairs, 2.5mm width, FFF linac) mounted on the robotic SRS/SBRT platform “CyberKnife M6™” in a pre‐clinical 5 months (11/2014–03/2015) test period. Methods: Beam properties were measured with unshielded diodes and EBT3 film. The CyberKnife workspace for MLC was analyzed by transforming robot node coordinates (cranial / body paths) into Euler geometry. Bayouth tests for leaf / bank position accuracy were performed in standard (A/P) and clinically relevant non‐standard positions, before and after exercising the MLC for 10+ minutes. Total system and delivery accuracy were assessed in End‐to‐End tests and dosimetric verification of exemplary plans. Stability over time was evaluated in Picket‐Fence‐and adapted Winston‐Lutz‐tests (AQA) for different collimator angles. Results: Penumbrae (80–20%, with 100%=2*dose at inflection point; SAD 80cm; 10cm depth) parallel / perpendicular to leaf motion were 2.87/2.64mm for the smallest (0×76×0.75cm2) and 5.34/4.94mm for the largest (9.76×9.75cm2) square field. MLC circular field penumbrae exceeded fixed cones by 10–20% (e.g. 60mm: 4.0 vs. 3.6mm; 20mm: 3.6 vs. 2.9mm). Interleaf leakage was <0.5%. Clinically accessible workspace with MLC covered (non‐coplanar) gantry angles of [‐113°;+112°] (cranial) and [‐108°;+102°] (body), and collimator angles of [‐100°;+107°] (cranial) and [‐91°;+100°] (body). Average leaf position offsets were ≤0.2mm in 14 standard A/P Bayouth tests and ≤0.6mm in 8 non‐standard direction tests. Pre‐test MLC exercise increased jaggedness (range ±0.3mm vs. ±0.5mm) and allowed to identify one malfunctioning leaf motor. Total system accuracy with MLC was 0.39±0.06mm in 6 End‐to‐End tests. Picket‐Fence and AQA showed no adverse trends during the test period. Conclusion: The Incise™ MLC for CyberKnife M6™ displayed high accuracy and mechanical stability over the test period. The specific CyberKnife geometry and performance after exercise demand dedicated QA measures. This work is in part funded by a research grant from Accuray Inc, Sunnyvale, USA. Erasmus MC Cancer Institute also has research collaborations with Elekta AB, Stockholm, Sweden. C Fuerweger has previously received speaker honoraria from Accuray Inc, Sunnyvale, USA.