Spine SBRT With Halcyon™: Plan Quality, Modulation Complexity, Delivery Accuracy, and Speed

Abstract

Purpose: Spine SBRT requires treatment plans with steep dose gradients and tight limits to the cord maximal dose. A new dual-layer staggered 1-cm MLC in Halcyon™ treatment platform has improved leakage, speed, and DLG compared to 120-Millennium (0.5-cm) and High-Definition (0.25-cm) MLCs in the TrueBeam platform. Halcyon™ 2.0 with SX2 MLC modulates fluence with the upper and lower MLCs, while in Halcyon™ 1.0 with SX1 only the lower MLC modulates the fluence and the upper MLC functions as a back-up jaw. We investigated the effects of four MLC designs on plan quality for spine SBRT treatments. Methods: 15 patients previously treated at our institution were re-planned according to the NRG-BR-002 guidelines with a prescription of 3,000 cGy in 3 fractions, 6xFFF, 800 MU/min, and 3-arc VMAT technique. Planning objectives were adjusted manually by an experienced planner to generate optimal plans and kept the same for different MLCs within the same platform. Results: All treatment plans were able to achieve adequate target coverage while meeting NRG-BR002 dosimetric constraints. Planning parameters were evaluated including: conformity index, homogeneity index, gradient measure, and global point dose maximum. Delivery accuracy, modulation complexity, and delivery time were also analyzed for all MLCs. Conclusion: The Halcyon™ dual-layer MLC can generate comparable and clinically equivalent spine SBRT plans to TrueBeam plans with less rapid dose fall-off and lower conformity. MLC width leaf can impact maximum dose to organs at risk and plan quality, but does not cause limitations in achieving acceptable plans for spine SBRT treatments.

Keywords: Halcyon™; SBRT; modulation complexity score; multi-leaf collimator; plan quality; spinal metastasis.

Figure 1

PTV Shapes—Examples of selected patients treated with spine SBRT. The CT slice displayed corresponds to the location of the cord dose maximum for Halcyon™ 1.0 generated treatment plans.

Figure 2

Cord Max Dose Comparison—Comparison of cord max dose or cauda equina (D0.03 cc) across all patients. Black dotted lines shown above correspond to the OAR dose limits for 3 fraction SBRT as defined in the NRG-BR002 protocol for spinal cord for a volume <0.03 cc. A spinal cord constraint for 3 fractions of 18-21 Gy D_max displayed as a blue dotted line (8). All plans have cord or cauda equina under the limit set in NRG-BR002. Maximum dose to spinal cord was found to have a range of [1,060–1,698] cGy for the Halcyon™ 2.0 with SX2, while the TrueBeams with Millennium MLC and HD MLC were found to have comparable maximum doses ranges of [1,006–1,688] cGy and [919–1,631] cGy, respectively.

Figure 3

Key Dosimetric Parameters—(A) Planning parameters are compared between the Halcyon™ and TrueBeam platforms to evaluate plan quality. (B) Matched paired analysis was performed comparing the difference between SX1, SX2, and Millennium-120 MLC to the High Definition-120 (HD) MLC for various plan parameters to evaluate statistically significant trends.

Figure 4

Treatment Delivery Parameters—Delivery parameter including total MU, modulation complexity score, and delivery time is compared for the Halcyon™ platform with SX1 and SX2 and TrueBeam platform with High Definition (HD) 120 MLC and Millennium 120 MLC. (A) The modulation complexity score is shown for Halcyon™ version 2 with SX2 is compared to TrueBeam Millennium MLC. (B) Shows the total MU delivered for each of the different treatment modalities. (C) All TrueBeam plans were adjusted to be 1,400 MU/min as compared to the 800 MU/min available in the Halcyon™ platform to utilize optimal delivery characteristics per treatment unit. Linear fits with the 95% confidence interval for delivery time compared to the total MU for all patients with the exception of case 3 due to exceptionally high MU.

Figure 5

Dose Fall-off inside Spinal Canal—(A) Dose profiles originating 1 cm from the edge of the PTV moving posteriorly toward the spinal cord are shown for each of the different treatment modalities for 4 patients. (B) Comparison of the isodose lines between various treatment platforms for case 2 are shown, thus displaying the conformality of the dose to the target volume as well as the fall off toward the spinal cord. PTV is shown as red translucent contour and likewise the spinal cord is shown as the magenta translucent contour.