Program Information

Feasibility Study of Using ABC to Gate Lung SBRT Treatment

D Cao*, X Xie , D Shepard , Swedish Cancer Institute, Seattle, WA

Presentations

SU-E-T-401 Sunday 3:00PM - 6:00PM Room: Exhibit Hall

Purpose:The current SBRT treatment techniques include free breathing (FB) SBRT and gated FB SBRT. Gated FB SBRT has smaller target and less lung toxicity with longer treatment time. The recent development of direct connectivity between the ABC and linac allowing for automated beam gating. In this study, we have examined the feasibility of using ABC system to gate the lung SBRT treatment.

Methods:A CIRS lung phantom with a 3cm sphere-insert and a moving chest plate was used in this study. Sinusoidal motion was used for the FB pattern. An ABC signal was imported to simulate breath holds. 4D-CT was taken in FB mode and average-intensity-projection (AIP) was used to create FB and 50% gated FB SBRT planning CT. A manually gated 3D CT scan was acquired for ABC gated SBRT planning.

An SBRT plan was created for each treatment option. A surface-mapping system was used for 50% gating and ABC system was used for ABC gating. A manually gated CBCT scan was also performed to verify setup.

Results:Among three options, the ABC gated plan has the smallest PTV of 35.94cc, which is 35% smaller comparing to that of the FB plan. Consequently, the V20 of the left lung reduced by 15% and 23% comparing to the 50% gated FB and FB plans, respectively. The FB plan took 4.7 minutes to deliver, while the 50% gated FB plan took 18.5 minutes. The ABC gated plan delivery took only 10.6 minutes. A stationary target with 3cm diameter was also obtained from the manually gated CBCT scan.

Conclusion:A strategy for ABC gated lung SBRT was developed. ABC gating can significantly reduce the lung toxicity while maintaining the target coverage. Comparing to the 50% gated FB SBRT, ABC gated treatment can also provide less lung toxicity as well as improved delivery efficiency.

Funding Support, Disclosures, and Conflict of Interest: This research is funded by Elekta.

Contact Email: