(Sunday, 7/14/2019) 3:30 PM - 4:00 PM
Room: Exhibit Hall | Forum 7
Purpose: Robust optimization(RO) is suggested in ICRU 91 to minimize the dose uncertainties so that a more consistent prescription scenario can be achieved. This study aims to evaluate the robustness(dose variations) under respiration utilizing RO and compared with standard optimization(SO) of traditional PTV approaches.
Methods: 14 SBRT lung cases were included in 4 planning strategies: (i)SO on the mean intensity projection generated from 4D-CT(3D_ITV) with a 5-mm isotropic expansion from internal target volume of maximum intensity project; (ii)SO on mid-ventilation phase with van Herk margin receipt(3D_VH); (iii)RO on mean intensity projection with the setup uncertainties derived from daily 4D-CBCT(3D_robust); (iv)RO on all 10 respiratory phases(4D_robust). All plans were then recalculated on the 10-phase data sets and finally deformed and accumulated into 4D doses. The doses of GTV and normal tissues under respiration and final 4D doses were compared among approaches and tested by Friedman and pair-wise Wilcoxon signed tests.
Results: The estimated total setup errors(RMS of Σ and σ) from 4D-CBCT used as RO uncertainties were 4.9mm(L/R), 6.9mm(A/P) and 4.3mm(S/I). The dose variations of normal tissues in all phases were also lower for RO group. Besides, 53 phases(7 patients), 35 phases(6 patients), 11 phases(2 patients) and 3 phases(1 patient) in 3D_ITV, 3D_VH, 3D_robust, 4D_robust respectively failed to comply with the chest wall tolerance. Adequate target coverage and significant lower chest wall dose(CW_V30Gy) and normal lung doses(NL_V20Gy, NL_V5Gy and MLD) were found in RO groups no matter in planning phases or 4D doses.
Conclusion: Uncertainty under respiration is an important issue in lung SBRT as the target/normal tissues might go into/out of the high and low fluence zones. Our study suggested that RO provides more robustness against respiratory motion and more consistent prescription for lung SBRT.
Keywords
Taxonomy
TH- External beam- photons: VMAT dose optimization algorithms
Contact Email
