Room: AAPM ePoster Library
Purpose: To investigate the dosimetry effect of cine magnetic resonance image (cMRI) gating in breath-hold pancreatic cancer radiotherapy.
Methods: This retrospective study was performed on 17 pancreatic cancer patients who received MRI guided stereotactic radiotherapy. Radiation was delivered during deep inspiration breath hold. The cMRI was used to real-time monitor tumor motion. A GTV tracking boundary of 3-5 mm was used. The radiation beam was turned off when >5% of the GTV moved out of the boundary. Because of the imaging-related latency, dose overshoot occurred when beam remained on while the tracking target is already moved out of the boundary. To investigate the dose overshoot and be more representative, we re-planned the 17 patients on a Linac machine using Pinnacle planning system with 33Gy prescription in 5 fractions. The patterns of tumor motion, particularly the target-out distance L and time ratio R, were extracted from the cMRI data. The isocenter was shifted by L to create an iso-shift plan representing radiation delivery at target-out. Then, the iso-shift plan was added to the original plan with a weight R to create a composite plan. PTV coverage and dose to nearby critical structures were compared between the composite and original plan.
Results: The target-out distance and time ratio are 7.0±2.0mm and 6.6±3.1%. The differences are significant for PTV V33Gy (93.7±1.8 vs 97.6±1.3%), duodenum V12.5Gy (9.5±1.0 vs 8.9±0.7cc) and V18Gy (3.3±0.8 vs 3.1±0.7cc, p<0.001 for all above parameters). The failure rates for PTV coverage, duodenum and stomach dose constraints are 82.4%, 41.2%, and 5.9%.
Conclusion: The dose overshoot caused by imaging gating latency significantly decreased the dose delivered to PTV, and increased the dose to immediate critical structures. It should be mitigated by implementing patient-initiated beam-on control to reduce unnecessary beam-on events and/or using faster MRI acquisition to reduce the latency.