Evaluation of kidney motion and target localization in abdominal SBRT patients

Abstract

The purpose of this study was to evaluate bilateral kidney and target translational/rotational intrafraction motion during stereotactic body radiation therapy treatment delivery of primary renal cell carcinoma and oligometastatic adrenal lesions for patients immobilized in the Elekta BodyFIX system. Bilateral kidney motion was assessed at midplane for 30 patients immobilized in a full-body dual-vacuum-cushion system with two patients immobilized via abdominal compression. Intrafraction motion was assessed for 15 patients using kilovoltage cone-beam computed tomography (kV-CBCT) datasets (n = 151) correlated to the planning CT. Patient positioning was corrected for translational and rotational misalign-ments using a robotic couch in six degrees of freedom if setup errors exceeded 1mm and 1°. Absolute bilateral kidney motion between inhale and exhale 4D CT imaging phases for left-right (LR), superior-inferior (SI), and anterior-posterior (AP) directions was 1.51 ± 1.00 mm, 8.10 ± 4.33 mm, and 3.08 ± 2.11 mm, respec-tively. Residual setup error determined across CBCT type (pretreatment, intrafrac-tion, and post-treatment) for x (LR), y (SI), and z (AP) translations was 0.63 ± 0.74 mm, 1.08± 1.38 mm, and 0.70 ± 1.00 mm; while for x (pitch), y (roll), and z (yaw) rotations was 0.24 ± 0.39°, 0.19 ± 0.34°, and 0.26 ± 0.43°, respectively. Targets were localized to within 2.1 mm and 0.8° 95% of the time. The frequency of misalignments in the y direction was significant (p < 0.05) when compared to the x and z directions with no significant difference in translations between IMRT and VMAT. This technique is robust using BodyFIX for patient immobilization and reproducible localization of kidney and adrenal targets and daily CBCT image guidance for correction of positional errors to maintain treatment accuracy.

Figure 1

Summary of translational and rotational positional errors based on 151 image registrations. The absolute errors are represented as a mean and standard deviation. The data were separated as: (a) x (LR), y (SI), and z (AP); (b) IMRT and VMAT; and (c) pre‐, intra‐, and post‐treatment CBCT. The * indicates statistical significance between the respective results at $p<0.05$.