A Dosimetric Comparison of the Accumulated Dose in Prostate SBRT for Non-Adaptive and Adaptive External Beam...

Background/Objectives: Traditional radiotherapy treatments assume that patient anatomy remains unchanged over the course of treatment. Image guidance is used to reproduce the patient setup as closely as possible, and planning margins are used to account for setup errors. With the development of MR-guided Adaptive Radiotherapy (MRgART), daily plan adaptations are now feasible, allowing clinicians to edit the plan according to daily anatomical fluctuations. However, MRgART is currently restricted to step-and-shoot IMRT delivery, which can have reduced dose conformality compared to VMAT. In this study, we compare the accumulated dose over all fractions in prostate SBRT treatments for non-adaptive and adaptive external beam workflows. Methods: The simulation and daily images of twenty previously treated MRgART prostate SBRT patients were anonymized. On each simulation image, whole prostate VMAT and MRgART SBRT plans were generated. To simulate non-adaptive treatment dose, the daily images were rigidly registered to the planning images, and the doses were recalculated on the daily images. The MRgART plans were adapted to the daily anatomy and reoptimized. All fractional doses were accumulated, using deformable image registration, and compared to their respective planned doses. Results: All MRgART dose accumulations were within clinical tolerance. Four VMAT dose accumulations had a dose constraint that fell outside of clinical tolerance. The VMAT dose accumulations had statistically lower doses to the target compared to their planned doses. Conclusions: While high-quality plans can be delivered in a non-adaptive VMAT workflow despite interfractional motion, this study suggests that MRgART produces cumulative dose distributions that more closely resemble the initial treatment plan.