Parameter Sensitivity Analysis of VMAT Planning for Spine SBRT
H Chen*, BA Winey, J Daartz, DP Gierga, Massachusetts General Hospital, Harvard Medical School, Boston, MASU-E-T-674 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
Purpose: To analyze the sensitivity of VMAT plan quality to various delivery parameters for spine SBRT.
Methods: Ten spine patients, previously treated to 18 Gy in 1 fraction with 11 field IMRT plans generated using a Multi-Criteria Optimization approach in Raystation (RaySearch Laboratories, Stockholm, Sweden) were selected for this study. Optimal VMAT plans, of comparable quality were generated using dual partial arcs and defined as baseline plans. The sensitivity of plan quality on dose rate level (variable dose rates (VDR), discrete dose rates (DDR)), leaf speed constraints (0.1, 0.2, 0.4, 0.6, 0.8 cm/degree and no constraint), was tested. The effect of maximum dose rate, maximum delivery time, and fraction size (18Gy*1, 9Gy*2 and 2Gy*9) was also tested. The resulting VMAT plan quality was analyzed by comparing PTV homogeneity index and conformity number, and D1 of spinal cord.
Results: A leaf speed constraint of less than 0.2 cm/degree will degrade PTV conformity and spinal cord sparing, and plan quality of leaf speed more than 0.6 cm/degree is close to that without any leaf speed constraint. The proper selection of VMAT delivery parameters allows VMAT plan quality to match IMRT. In addition, VMAT plan quality of VDR is superior to that of DDR when either maximum delivery time or maximum dose rate is increased.
Conclusion: VMAT plan quality for spine SBRT depends on the combination of delivery time, fraction size, and leaf speed. The sensitivity of plan quality depends on dose per fraction and maximum delivery time.
Funding Support, Disclosures, and Conflict of Interest: Funding support: NCI Federal Share Proton Beam Program Income Grant. Conflict of Interest:none.