Exhibit Hall | Forum 7
Purpose: To investigate the feasibility of thoracic spine SBRT in retreatment setting on Elekta Unity MR-Linac. We hypothesized that Monaco planning can improve the GTV minimum dose (GTVmin) with comparable target coverage and maximum spinal cord dose and can maintain consistent plan quality during the daily adaptation process.
Methods: Reference plans were regenerated in the Monaco planning system for the Unity MR-Linac using 9 and 13 IMRT beams for 10 thoracic spine SBRT cases. For each reference plan, adapt-to-position (ATP) and adapt-to-shape (ATS) plans were generated using MRIs with simulated daily positional setup error. Monaco plans were compared to clinical Pinnacle plans, and the adaptive plans were compared with Monaco reference plans. The plan quality index included target coverage, Paddick conformity index (PCI), gradient index, homogeneity index, spinal cord and esophagus D0.01cc, lung V10Gy, and skin D0.01cc.
Results: GTVmin from Monaco 9-beam and 13-beam plans are significantly higher than Pinnacle plans (p<0.01). The GTV PCI are comparable among the three plans, while the CTV PCI from the Monaco 13-beam plans are significantly higher than Pinnacle (p<0.02) and Monaco 9-beam plans (p<0.01). No statistically significant difference between the 3 plans was seen for the spinal cord D0.01cc, esophagus D0.01cc or lung V10Gy. The electron-return-effect did not induce remarkable dose effects around the lungs, esophagus or skin. In the ATS workflow, GTVmin was consistently within 5% of the reference plan. In the ATP workflow, a dramatic increase in GTVmin was observed at the cost of > 10% increase in the cord D0.01cc.
Conclusion: MRL plans for thoracic spine SBRT demonstrated safety and feasibility of tumor dose escalation with cord dose preservation and stability of plan adaptation using the ATS workflow. Careful plan review of hot spots, lung dose, and other low dose volumes is necessary for a safe MRL treatment.