Unencrypted login | home

Program Information

Dosimetric Effects of Diaphragm Motion On Spine SBRT Treatments


B Winey

B Winey*, H Chen, J Daartz, D Gierga, J Shin, J Schwab, K Oh, Massachusetts General Hospital and Harvard Medical School, Boston, MA

SU-E-J-207 Sunday 3:00PM - 6:00PM Room: Exhibit Hall

Purpose: To evaluate the dosimetric effects of diaphragm motion on spine SBRT treatments with IMRT and VMAT deliveries.

Methods: Five 4D-CTs were acquired for lung and liver SBRT patients. The vertebral body closest to the diaphragm level was contoured as the CTV. The thecal sac was used as a surrogate for the spinal cord PRV. A 3 mm expansion was used for the PTV with the PTV cropped to the cord PRV. Plans were optimized for PTV maximal coverage at 1800 cGy and cord PRV Dmax was no more than 1200 cGy. Plans were created for 7 and 11 field IMRT and partial arc VMAT modalities, all on the average intensity CT (AvgCT) and recalculated on the 4D CT data set. D95, D1, and Davg were recorded for the different delivery methods and all 11 CT sets (AvgCT+10 phases). Dose differences were calculated for each phase versus the planning AvgCT and for the maximum difference amongst the phases to evaluate worst-case scenarios.

Results: Regarding the PTV coverage, the IMRT-7 plans always exhibited the largest dosimetric differences amongst metrics with Davg having the largest range (Davg[max-min]: 63 cGy) versus the IMRT-11 (51 cGy) and VMAT (49 cGy) out of the prescription dose of 1800 cGy. All other dosimetric points had smaller differences for PTV metrics. The maximum difference always occurred between exhale and inhale phases. The cord PRV D1% had smaller differences than PTV metrics with the maximum dose difference of 44 cGy for the IMRT-7 plans, and IMRT-11 had a maximum of 17 cGy and VMAT 23 cGy.

Conclusions: Clinically relevant spine SBRT plans display less than 4% maximum difference in PTV dose (Davg) and 3% variability in the cord PRV D1% metric between inhale and exhale phases for IMRT-7 plans in the worst case scenario.

Contact Email: