Room: Karl Dean Ballroom A2
Purpose: To demonstrate the feasibility of using synthetic CT of pelvis generated from daily image-guided CBCT for monitoring changes in water equivalent path length (WEPL) during the course of proton therapy and determine the dosimetric impact of anatomy changes along the proton beam path.
Methods: Ten representative pediatric patients with pelvic tumors treated using proton therapy with daily CBCT were included. On the days when repeat CT was performed for adaptive replanning, initial planning CT was deformed to the same-day CBCT in MIM software (normalized intensity-based deformable image registration) to generate synthetic CT images for WEPL comparison and dosimetric evaluation. WEPL differences of actual treatment fields were calculated for the pairs of (initial CT - repeat CT) and (initial CT - synthetic CT) of the day using in-house software. Any deviation from the original plan was investigated and dosimetric impact estimated using the Eclipse treatment planning system.
Results: WEPL changes of 20 proton fields at the distal edge of the CTV ranged from 0.1-12mm with a median=2.5mm, and 75th percentile=5.1mm for (initial CT - repeat CT), and ranged from 0.3-10.1mm with a median=2.45mm, and 75th percentile=4.8mm for (initial CT - synthetic CT). Statistical analysis on WEPL differences comparing (initial CT - repeat CT) and (initial CT - synthetic CT) revealed that the two methods were strongly correlated (r=0.93) and showed no statistically significant difference (p=0.81, Î±=0.05). The mean difference(s) were/was -0.09Â±1.65mm. Dosimetric impact was due to proton range pullback, which led to reduced coverage (CTV Dmin) averaging 12.1% and 11.3% in repeat and synthetic CT verification plans, respectively.
Conclusion: The study demonstrated that synthetic CT generated by deforming the initial CT to CBCT of the day is reasonably accurate to monitor proton range variations in the pelvis and clinically feasible to eliminate excessive CT scans for pediatric adaptive proton therapy.