MRI-Guided Adaptive Head and Neck Radiotherapy On a 1.5T MR Linac System
E Aliotta, J Hesse, E Subashi, P Godoy Scripes, M Aristophanous*, J Mechalakos, L Cervino, N Tyagi, K Zakeri, N Lee, Memorial Sloan Kettering Cancer Center, New York, NY
Presentations
PO-GePV-M-242 (Sunday, 7/10/2022) [Eastern Time (GMT-4)]
ePoster Forums
Purpose: Adaptive head and neck (H&N) radiotherapy can preserve treatment plan quality and spare organs-at-risk (OAR) in the presence of changing anatomy. MR-guided linear accelerators (MR-Linac) permit frequent plan adaptations with high-quality, non-ionizing on-board imaging. In this study, we report on H&N patients treated on MR-Linac with weekly adaptation and assess the dosimetric benefits of this strategy.
Methods: Patients received conventionally fractionated H&N radiotherapy on a 1.5T MR-Linac. Online-adaptation (adapt-to-shape) was performed weekly, which included deformable contour propagation, manual contour modification, and plan re-optimization. Treatment times (patient setup to beam-off) were recorded for each fraction. Delivered adaptive doses were assessed by summing relevant plan-quality-metrics (OAR Dmean/Dmax, target D95%) from each plan (scaled by the number of fractions delivered between adaptations). Hypothetical non-adaptive doses were assessed through deformable forward propagation wherein initial planned dose maps were deformed to subsequent MRIs to estimate the dose that would have been delivered (without adaptation) on that day. Cumulative non-adaptive doses were determined by summing plan-quality-metrics across timepoints. Adaptive and non-adaptive target/OAR doses were then compared against each-other and against initial plan dosimetry.
Results: N=7 patients were treated. In total, 223 fractions were delivered, including 41 adaptations (18% of fractions). Excluding first-fraction patient setups, average treatment times were 39±10 minutes for non-adaptive (range: 19-79), and 66±19 minutes for adaptive fractions (range: 31-105). Dose accumulation showed that without adaptation, PTV D95% would have decreased significantly compared to baseline dosimetry (ΔD95%=-3.1±4.2%, P=0.02). However, delivered adaptive D95% did not differ from initial plans (ΔD95%=+0.3±0.6%, P=0.24) and increased significantly compared to the non-adaptive scenario (ΔD95%=+3.7±4.8%, P=0.005). Adaptive OAR doses were generally lower than non-adaptive (ΔDmean/Dmax=-37.6±139.4cGy, P=0.08), but differences were only statistically significant for spinal-cord (ΔDmax=-106.1±146.8 cGy, P=0.006).
Conclusion: H&N radiotherapy with weekly adaptation was performed on a 1.5T MR Linac. Adaptation improved target coverage compared to simulated non-adaptive treatment.
Keywords
Taxonomy
IM/TH- MRI in Radiation Therapy: MRI/Linear accelerator combined (general)
Contact Email