PV QA 3 - Poster Viewing Q&A 3
Purpose/Objective(s): We prospectively assessed for the localization accuracy of the three-degrees-of-freedom (3-DOF) fiducial markers (FM)-based setup without rotational correction using an oblique 2D X-ray imagers (OXI) for patients with prostate cancer receiving hypofractionated intensity modulated radiotherapy (IMRT).
Materials/Methods: We analyzed 22 consecutive patients with prostate cancer with three implanted FMs who underwent hypofractionated IMRT (70 Gy in 28 fr. or 62 Gy in 20 fr.) at our hospital between August 2015 when we started the 3-DOF-FM-based setup and March 2017 were selected in this study. At treatment, a patient was initially positioned to the planning CT isocenter based on skin markers with a laser coordination system. After the initial setup, the automatic 6-DOF pelvic bony-structure-based matching using the OXI and then the couch shift was carried out. Subsequently, the automatic 3-DOF-FM-based image matching using the OXI and then the couch shift was carried out. After that, for the purpose of reviewing the soft tissue (ST)-based matching, cone-beam computed tomography (CBCT) images were obtained continuously for four days after the start of treatment, and thereafter, those were obtained once a week. After the image matching was reviewed and approved by radiation oncologists, the radiation was delivered. After treatment, the OXI images were obtained and the intra-fractional prostate shifts were reported. The residual error of the 3-DOF-FM-based setup respect to the ST-based matching using CBCT and correlation between CBCT and OXI shifts were estimated. The rotational mean shifts detected by OXI were also evaluated. In addition, the PTV margins in FM-based setup by Van Herk’s recipe were calculated from pre and post treatment OXI images.
Results: A total of 176 image datasets were analyzed. On the mean residual error of the 3-DOF-FM-based setup in the anterior-posterior (AP), superior-inferior (SI), and left-right (LR) dimensions were 0.57 ± 1.58 mm, -0.21 ± 1.31 mm, -0.01 ± 0.59 mm. The percentages of agreements within ±2 mm were 80% for AP, 94% for SI, 100% for LR. The Pearson coefficient of correlation between both shifts were 0.96 for AP and SI, 0.78 for LR. The rotational mean shifts were -0.23 ± 1.14° for AP, -0.61 ± 1.52° for SI, 0.35 ± 4.26° for LR. The PTV margins were 3.8 mm for AP, 3.0 mm for SI, 1.2 mm for LR.
Conclusion: In prostate hypofractionated IMRT, the 3-DOF-FM-based setup without the rotational correction has a small residual error for ST-based matching using CBCT and there was a high correlation between them, so that it can reduce setup error and provide the accuracy comparable to ST-based setup. In addition, it is expected to shorten treatment time and reduce intra-fractional prostate shift.
The asset you are trying to access is locked. Please enter your access key to unlock.