Setup reproducibility in radiation therapy for lung cancer: a comparison between T-bar and expanded foam immobilization devices

Abstract

Purpose: Physiologic and non-physiologic tumor motion complicates the use of tight margins in three-dimensional (3D) conformal radiotherapy. Setup reproducibility is an important non-physiologic cause of tumor motion. The objective of this study is to evaluate and compare patient setup reproducibility using the reusable T-bar and the disposable expanded foam immobilization device (EFID) in radiation therapy for lung cancer. Methods and Materials: Two hundred forty-four portal films were taken from 16 prospectively accrued patients treated for lung cancer. Patients were treated with either a pair of anterior and posterior parallel opposing fields (POF), or a combination of POF and a three-field isocentric technique. Each patient was treated in a supine position using either the T-bar setup or EFID. Six patients were treated in both devices over their treatment courses. Field placement analysis was used to evaluate 3D setup reproducibility, by comparing positions of bony landmarks relative to the radiation field edges in digitized simulator and portal images. Anterior-posterior, lateral, and longitudinal displacements, as well as field rotations along coronal and sagittal planes were measured. Statistical analyses of variance were applied to the deviations among portal films of all patients and the subgroup treated with both immobilization methods. Results: For the T-bar immobilization device, standard deviations of the setup reproducibility were 5.1, 3.7, and 5.1 mm in the anterior-posterior, lateral, and longitudinal dimensions, respectively. Rotations in the coronal plane and the sagittal plane were 0.9° and 1.0°, respectively. For the EFID, corresponding standard deviations of set up reproducibility were 3.6 mm, 5.3 mm, 5.4 mm, 0.7° and 1.4°, respectively. There was no statistically significant difference ( p = 0.22) in the 3D setup reproducibility between T-bar and EFID. Subgroup analysis for the patients who were treated with both immobilization devices did not reveal a difference either. There was no consistent systematic error from simulator to treatment unit identified for either immobilization device. Conclusion: Although the optimal immobilization technique and patient positioning for thoracic radiotherapy have yet to be determined, this study indicates that T-bar is comparable with EFID in its setup reproducibility. In view of the inherent advantages of T-bar, it has become a standard immobilization device at our institution. The observed range of displacements in field positioning with either immobilization device implies that one cm (two standard deviations [SD] of setup error) will be a more appropriate margin to allow for setup variability in radiation therapy for lung cancer.