Room: AAPM ePoster Library
Purpose: To assess differences in daily accumulated dose for robustly optimized intensity-modulated proton therapy (RO-IMPT) plans using different uncertainty scenarios.
Methods: Eight prostate cancer patients previously treated with VMAT and having daily CBCT scans were included in this study. First, RO-IMPT plans were created with ±3mm and ±5mm patient setup uncertainties and ±3% proton range uncertainties. Second, the planning CT (pCT) was deformably registered to daily CBCTs to create synthetic CTs (sCT) by transferring the electron densities from pCT to CBCTs. Doses were recalculated and evaluated on daily sCTs for both ±3mm/±3% and ±5mm/±3% uncertainties, and were accumulated back to the pCT. Accumulated doses generated from ±3mm/±3% and ±5mm/±3% RO-IMPT plans were assessed and compared using the clinical dose volume constraints for CTV, bladder and rectum.
Results: Daily accumulated dose based on ±3mm/±3% and ±5mm/±3% uncertainties for RO-IMPT plans resulted in satisfactory CTV coverage (CTV(V95%)= 98.85 ± 1.20% vs. CTV(V95%)=99.79 ± 0.21%, P<0.01). However, the accumulated dose based on ±3mm/3% RO-IMPT plans consistently provided greater OAR sparing than ±5mm/3% RO-IMPT plans (Rectum(V65) = 2.56 ± 1.87% vs. 4.15 ± 2.71%, P< 0.01; Bladder(V65)= 4.96 ± 7.96% vs. 6.88 ± 10.80%, P< 0.01). Evaluation of daily recalculated dose on sCTs showed that higher percentage of fractions of the ±5mm/±3% RO-IMPT plans satisfied the CTV(V95%)> 95% than the ±3mm/±3% RO-IMPT plans did (100% vs. 95.9%). However, much higher percentage of daily fractions satisfied the CTV(Dmax)< 115% (99.7% vs. 91.8%) and Bladder(V65)< 25% (91.4% vs. 88.0%) criteria using ±3mm/3% RO-IMPT plans.
Conclusion: This study demonstrated that ±3mm/3% uncertainty scenario for robust IMPT optimization is sufficient to meet CTV coverage constraint while maintaining lower doses to OARs than ±5mm/3% plans. Furthermore, the daily dose fluctuations do not seem to exceed the dose constraints for majority of the fractions.