Discrepancy Between the Intended Isocenter and That from Orthogonal Linacgrams in the Treatment of CT-guided Stereotactic Body Radiotherapy(SBRT)

PURPOSE: In our institution a CT scanner was installed in the same room as the linear accelerator. In stereotactic body radiotherapy (SBRT) we confirmed the isocenter position by serial thin-slice and long-scan-time CT images before every treatment as well as in planning. In planning we constructed digitally reconstructed radiography (DRR) of both the anterior and lateral views. At the first treatment we also checked the isocenter with linacgraphy. Then we compared the isocenter positions obtained from the DRR and linacgraphy. MATERIALS AND METHODS: Between Feb. 2005 and Oct. 2006, we treated 75 lung and liver tumors with SBRT in this way. Based on bony structures, we measured the differences between in-isocenter positions for SI, LR, and AP directions between DRR and linacgraphy. RESULTS: The median (min-max) of the differences in-isocenter positions for SI, LR, and AP directions between DRR and linacgraphy were 0.0 mm (0-6.0), 0.0 mm (0-10.0), and 0.0 mm (0-10.0), respectively, as well as 3.2 mm (0-12.3) for 3-dimensional distance. In 28 tumors (37%) the differences exceeded 5 mm in three-dimensional distance. The frequency of differences exceeding 5 mm in upper lung lesions tended to be more than that in liver lesions, and that in left pulmonary lesions was significantly more than that in right ones. CONCLUSION: This result suggests that the relative position of the target volume to the bony structure differ in planning and in every treatment. It was recommended to verify isocenter accuracy in institutions where isocenter position is checked only by orthogonal linacgraphy in SBRT.