A novel approach to SBRT patient quality assurance using EPID-based real-time transit dosimetry

Purpose Intra- and inter-fraction organ motion is a major concern in stereotactic body radiation therapy (SBRT). It may cause substantial differences between the planned and delivered dose distribution. Such delivery errors may lead to medical harm and reduce life expectancy for patients. The project presented here investigates and improves a rapid method to detect such errors by performing online dose verification through the analysis of electronic portal imaging device (EPID) images. Methods To validate the method, a respiratory phantom with inhomogeneous insert was examined under various scenarios: no-error and error-simulated measurements. Simulation of respiratory motions was practiced for target ranges up to 2 cm. Three types of treatment planning technique – 3DCRT (three-dimensional conformal radiation therapy), IMRT (intensity modulated radiation therapy), and VMAT (volumetric modulated arc therapy – were generated for lung SBRT. A total of 54 plans were generated to assess the influence of techniques on the performance of portal dose images. Subsequently, EPID images of 52 SBRT patients were verified. Both for phantom and patient cases, dose distributions were compared using the gamma index method according to analysis protocols in the target volume. Results The comparison of error-introduced EPID-measured images to reference images showed no significant differences with 3%/3 mm gamma evaluation, though target coverage was strongly underestimated. Gamma tolerance of 2%/2 mm reported noticeable detection in EPID sensitivity for simulated errors in 3DCRT and IMRT techniques. The passing rates for 3DCRT, IMRT, and VMAT with 1%/1 mm in open field were 84.86%, 92.91%, and 98.75%, and by considering MLC-CIAO + 1 cm (threshold 5%), were 68.25%, 83.19%, and 95.29%, respectively. Conclusion This study demonstrates the feasibility of EPID for detecting the interplay effects. We recommend using thin computed tomography slices and adding sufficient tumor margin in order to limit the dosimetric organ motion in hypofractionated irradiation with preserved plan quality. In the presence of respiratory and gastrointestinal motion, tighter criteria and consequently using local gamma evaluation should be considered, especially for VMAT. This methodology offers a substantial step forward in in vivo dosimetry and the potential to distinguish errors depending on the gamma tolerances. Thus, the approach/prototype provides a fast and easy quality assurance procedure for treatment delivery verification.