Purpose: To investigate the dosimetry effects of gating threshold selection in cine-MRI guided breath-hold pancreatic cancer radiotherapy.
Methods: Seventeen pancreatic cancer patients underwent MRI guided breath-hold stereotactic body radiation therapy. The sagittal cine-MRI was used to monitor tumor motion and gate radiation beam. Gross tumor volume (GTV) was set as the tracking target, and a margin of 3-5 mm was used as gating boundary. Beam-off event would not be triggered until the target-out percentage exceeded a predefined gating threshold. We re-generated VMAT plans using 33Gy prescription. We calculated the GTV displacement in each MRI frame that satisfied the gating threshold, shifted the isocenter by the displacement, and then added up all iso-shift plans to create the motion plan. The coverage for GTV and PTV and dose to nearby organs at risk (OARs) were compared between motion plans and original plan.
Results: The GTV V33Gy was 89.34%±12.58% in the original plan, and 88.53%±12.27%, 88.39%±12.16%, 88.30%±11.91% and 88.26%±11.82% for motion plans generated with 0%, 1%, 3%and 5% gating threshold. (p>0.05 for all motion plans). The PTV V33Gy was 97.63%±1.33% in the original plan, and 92.87%±3.16%, 92.32%±3.27%, 91.64%±3.38% and 91.27%±3.34% for motion plans generated with 0%, 1%, 3%and 5% gating threshold. (p<0.05 for all motion plans). The duodenum V12.5Gy (p<0.05) and V18Gy in the motion plan increased significantly when gating threshold was greater than 2% (p<0.05). No significant difference was found for other OARs. No significant difference was found for all parameters between motion plans.
Conclusion: The dose parameters deteriorate with increasing gating threshold but rendered no difference among thresholds 0-5%. But the efficiency gradually improves with increasing thresholds. A threshold of 3-5% was deemed a reasonable selection considering the tradeoff between effectiveness and efficiency.