Feasibility of Texture Analysis–Based Dosimetry in Computed Tomography Scans During Radiation Therapy of Non-Small Cell Lung Cancer

Abstract

Both respiratory-gated volumetric modulated arc radiation therapy (VMAT) and tumor-tracking robotic radiation therapy (RRT) are accurate and effective treatment options for liver cancer stereotactic body radiation therapy (SBRT). The purpose of this study was to assess dosimetric benefits facilitated with RRT's tumor tracking capability, where internal target volume (ITV) would not be necessary, compared to linear acceleratorebased gated treatment. Materials/Methods: A total of 29 hepatocellular carcinoma (HCC) patients previously treated using respiratory-gated VMAT (ITV, 30%-70% phases) with 2 arcs were selected for this study. In the original treated VMAT plan, individual ITV margin was added around gross tumor volume (GTV) by measuring GTV motion with 4DCT, followed by 5-mm isotropic planning target volume (PTV) margin. For each VMAT plan, 2 comparison RRT plans were generated using the original (RRT original , ITV included) and modified PTV (RRT modified , ITV excluded), respectively. Planning objectives were to deliver 45 Gy in 3 fractions to PTV >95% and conformity index (CI) <1.2, and to minimize doses to critical structures (esophagus, large bowel, stomach, duodenum, and spinal cord). In each case, the RRT original and RRT modified plans were compared to the corresponding VMAT plan in terms of dosimetric parameters: conformity index (CI), PTV coverage (CO), organs-at-risk (OAR) doses, and normal liver volume sparing. Results: The original PTV volume with median 24 mL (range 9-65 mL) was significantly reduced to median 12 mL (range 5-41 mL) for the RRT modified plan by excluding ITV margin. Statistical significant differences in plan qualities were observed between the VMAT and RRT plans, CI: mean 1.0 in VMAT vs. 1.2 in both RRTs (p<0.001), and CO: mean 93.0% in VMAT versus 96.6% in RRT original versus 96.9 in RRT modified (p<0.001) within the planning objectives. The dose falloff or intermediate-dose spillage characterized by R 50% was not statistically different. Although all the critical organ doses were well within tolerance dose-volume limit in all plans, the averaged normal liver volume receiving more than 5-45 Gy were significantly decreased in RRT modified than in VMAT and RRT original by 1.39 and 1.50 folds, respectively. Especially, normal liver volume (GTV exclude) receiving more than 15 Gy was significantly more decreased in RRT modified than in VMAT and in RRT original by 1.75 and 1.61 folds, respectively. Conclusion: The tumor-tracking capability of an RRT system can significantly decrease normal liver volume receiving more than 15 Gy, while maintaining high precision in target localization, conformity, tumor coverage, and the sparing of the OAR. Thus it would be a valuable treatment option for HCC patients with poor liver function.