Session Item

Stereostatic Body Radiotherapy (SBRT) of lung cancer is usually deliver a high fraction dose to a moving target with heterogeneous density gradient. Strategies in considering setup and dose estimation uncertainties are particularly important. This study aimed to compare the conventional margin-based target volume planning with the robust optimization planning approaches in terms of plan qualities in SBRT of centrally located non-small cell lung cancer (CL-NSCLC).

Twelve CT sets for SBRT of CL-NSCLC were recruited retrospectively. Two SBRT plans will be recomputed by conventional optimization with Planning Target Volume (PTV-O) and the Robust Optimization with the Internal Target Volume (ITV-RO) for each CT set. ITV was created by merging Gross Target Volume (GTV) of 10 motion phases of 4DCT. While 5mm isotropical margin was given to the ITV to form the PTV. The prescribed scheme was 50 Gy in 5 fractions. For comparison purposes, the prescribed dose was normalized to ITV after PTV-O. Planning goals were to cover 95% ITV by varying prescription isodose lines at 70 to 90% in both plans. 5mm isotropic uncertainties with +/- 2% density error (16 scenarios) was introduced to robustness analysis for both optimization approaches. The bandwidth measurement at D95% of ITV from DVH were measured for comparing the robustness. Moreover, worst-case scenarios of dosimetric parameters e.g., ITV coverage, dose spillage (R50%, the ratio of 25Gy isodose volume to ITV) and doses to organs at risk (OARs: lungs, spinal cord, heart, esophagus, trachea and major vessels) were also compared.

Similar target coverage was observed in all ITV parameters by both approaches but better performance was noted in ITV-RO approach: conformity of ITV (conformity index: 1.13 vs 1.21, p < 0.05), dose spillage (R50% ratio 4.18 vs 4.45, p < 0.05). The worst-case scenarios OARs parameters by ITV-RO were generally lower than that by PTV-O, significant differences were recorded in maximum dose of heart, major vessels and V20Gy (Lungs). The robustness of ITV-RO approach was also assured, the bandwidth of D95% of ITV was 39.6% narrower than that of PTV-O approach (p <  0.01).

ITV-RO approach offering advantages in conformity, spillage control and OARs sparing while without sacrificing the target coverage. Moreover, the robustness of ITV-RO also demonstrated superiority of withstanding setup and dose estimation uncertainties. Overall, it has potential clinical benefit to SBRT planning of CL-NSCLC.