Room: Exhibit Hall | Forum 7
Purpose: To evaluate the MR-Linac treatment planning system (TPS) for IMRT-based spine SBRT planning, and compare to clinical experience using VMAT planning.
Methods: Eight patients who underwent spine SBRT planned using VMAT delivery in 3-10 fractions (30-50Gy Rx dose), spanning a range of PTV sizes (25-160 cc) and geometries were re-planned using the MR-Linac TPS. The spinal cord (+1mm margin PRV) and esophagus were the primary OARs considered. Plan constraints were introduced first for PTV coverage, then maximum PTV dose, followed by OAR DVH limits, and maximum OAR dose. Constraints were matched to the planning goals used for the clinical VMAT plans, and depended on the location and fractionation. Finally, a cost function was used to improve conformity and reduce entrance dose. In both the clinical and MR-Linac plans PTV coverage was reduced if the spinal cord PRV goals could not be achieved. Optimization was performed at 2mm resolution and recalculated at 1mm to match the clinical plans. The MR-Linac and VMAT plans were compared dosimetrically and for delivery time.
Results: The resulting coverage (Vâ‚?â‚€â‚€) for the clinical plans (81.2%-98.9%) was not significantly different from the MR-Linac plans (85.9%-95.9%). The lower isodose coverage however (Vâ‚‰â‚€ and Vâ‚‰â‚…) was significantly better in the clinical plans (97+/-1% and 98+/-1% vs 94+/-1% and 96+/-1% respectively). The max dose was higher for the MR-Linac (134+/-3% vs. 120+/-2%), and the conformity was worse (1.27+/-0.06 vs. 1.06+/-0.06). The spinal cord PRV sparing trended better for the MR-Linac plans (Dmax 20+/-3Gy vs. 24+/-2Gy). Beam on time for the VMAT plans was significantly lower than the MR-linac plans (9.6+/-1.2mins vs. 12.7+/-1.3mins).
Conclusion: IMRT based SBRT planning using the MR-Linac TPS was feasible, however a higher degree of plan heterogeneity (i.e. max dose to PTV) was required to achieve the same PTV coverage and OAR sparing.