Comparison of liver stereotactic body radiotherapy plans based on free breathing and averaged CTs after trans‐arterial chemoembolization using lipiodol

Purpose To identify which CT set is more suitable for predicting the four‐dimensional (4D) cumulative dose distribution in liver stereotactic body radiotherapy (SBRT) with lipiodol retention. Methods For 15 patients, who underwent liver SBRT after trans‐arterial chemoembolization (TACE), volumetric modulated arc therapy (VMAT) plans were retrospectively generated on free breathing (FB) and average intensity projection (AVG) CT sets using the x‐ray Voxel Monte Carlo (XVMC) algorithm. The three‐dimensional (3D) FB and AVG plans were compared with corresponding 4D plans, which were created by deformable registration using 10‐phase plans. For both internal target volume (ITV) and planning target volume (PTV), the volume covered by the prescription dose of 36 Gy (V36) and the dose covering 95% of the volume (D95) were evaluated. The relative (3D/4D) normal liver volume >19.2 Gy and rib volume 28.8 Gy were also compared for FB and AVG plans. Results The mean values of the relative electron densities inside the PTV for FB and AVG were slightly different (1.050 for FB vs. 1.039 for AVG). For both ITV and PTV, V36 and D95 showed no statistically significant difference between FB and AVG. Also, the relative rib volumes >28.8 Gy and the normal liver volumes >19.2 Gy were not statistically different between FB and AVG plans. Conclusion For liver SBRT with lipiodol retention, 3D plans based on FB and AVG images were virtually equivalent in producing the 4D cumulative doses under the accuracy of the XVMC algorithm.