Choice of appropriate beam model and gantry rotational angle for low-dose gradient-based craniospinal irradiation using volumetric-modulated arc therapy

Abstract

AbstractObjectivesWe aimed to assess the impact of advanced multileaf collimator (MLC) models and flattening filter-free (3F) beam in volumetric-modulated arc therapy (VMAT)-based craniospinal irradiation (CSI).MethodsCT scans of five medulloblastoma patients who previously received CSI at our hospital were used for the present study. Patients were planned for a prescription dose of 35 Gy to craniospinal axis. A three-dimensional conformal radiotherapy (3DCRT) plan and a VMAT plan using 1 cm MLC leaf width were generated as the gold standard (reference arm). Test VMAT plans were generated using Agility MLC model (MLC leaf width 5 mm) for various combinations of flattened beam (F) and 3F beam for treating the brain and spine planning target volume (PTV). Organs at risks (OARs) were analysed for dose 5, 50, 75 and 90% volumes, mean dose and maximum dose.ResultsAll 3DCRT plans and VMAT plans were aimed to cover 95% of PTV by at least 95% prescription dose. VMAT demonstrated lesser dose spillage than 3DCRT to body volume minus PTV (NTID: non tumor integral dose) for a dose threshold above 7·5 Gy. For the low-dose range (1–7 Gy), variation between the dose coverage between all VMAT plans (for either spine or brain PTV) was <1%. Intra-VMAT plan dose variation for all OAR’s for all tested parameters was <1 Gy. Average monitor unit (MU) difference among different VMAT plans ranged between 1·52 and 2·13 when normalised to 3DCRT MU. For VMAT plans, flat beam with 1 cm MLC showed the highest MU, whereas Agility MLC with 3F beam had the least MU values for intra-VMAT plans. No statistical significance variation (p) was observed in between reference arm and test arm plans except for mean dose and V107% for PTV spine. When compared between reference arm 3DCRT and test arm VMAT plans. For OAR’s, no statistical difference was observed between reference and test arm VMAT plans.ConclusionsReference arm plans and test arm plans exhibit no statistically significant difference. However, as compared with 3DCRT, VMAT plans are more conformal and produce lesser dose to OAR at the cost of higher delivered MU. 3F beams or finer width MLC’s (width <5 mm) have no advantage over the conventional 1 cm MLC and flat beam except that 3F beams have a shorter beam delivery time. This study demonstrate a significantly lesser spillage dose to NTID/body that of the reported literature, which is attributed to limited rotational arc length used for VMAT plans.