MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel

Quantification of Prostate Movement with On-Demand KV Imaging During SBRT

K C Jones*, J Paul , Z Grelewicz , N Sen , D Wang , J Turian , Rush University Medical Center, Chicago, IL

Presentations

(Tuesday, 7/31/2018) 7:30 AM - 9:30 AM

Room: Room 205

Purpose: To assess the benefit of kV intrafraction on-demand imaging during prostate SBRT.

Methods: Triggered on-board kV images were collected at 70 MU intervals during Truebeam (Varian) delivery of ≤5 fraction SBRT prostate treatments. As per clinical practice, operators interrupted treatment if the real-time, auto-identified positions of the three implanted fiducials strayed more than 3 mm from their planning positions. Intra- or inter-arc interruption instigated cone beam CT (CBCT) and patient realignment before delivery resumption. The 2153 intra-delivery kV images collected for 15 patients were retrospectively analyzed. The fiducial positions were auto-identified with a template tracking algorithm before human verification. Here, our study focuses on analyzing sup-inf displacement.

Results: Despite low contrast at lateral angles, 82.7% of the fiducials were identified on the kV images. Of the investigated 50 fractions, treatment was interrupted as a result of kV fiducial tracking for 10 of the 15 patients a total of 16 times. Interruption, re-CBCT, and resulting couch re-positioning extended average inter-beam pauses from 39 to 226 s. Root-mean-squared couch shifts as a result of re-CBCT were 2.1, 2.4, and 0.9 mm for vertical, longitudinal, and lateral directions, respectively (maximum shift of 4.2 mm, longitudinal). Negligible systematic movement was observed. For identifiable fiducials, the sup-inf fiducial displacement was outside of the 3 mm threshold in 4% of cases. Subtracting applied couch shifts shows that without interruptions and re-positioning, 9.8% of fiducials would have been >3 mm from their planned sup-inf positions. In the most extreme patient case, absence of on-demand IGRT would have reduced the sup-inf 3 mm compliance from 83 to 47%.

Conclusion: There is substantial (>3 mm) prostate and/or patient movement during SBRT treatment even with extensive immobilization. Although each interruption extends overall treatment by 3 minutes, patient realignment triggered by kV on-demand imaging led to improved target positioning.

Keywords

Target Localization, Image-guided Therapy

Taxonomy

IM/TH- RT X-ray Imaging: General (most aspects)

Contact Email