Dose escalation in rectal radiotherapy shows promise for improving complete response rates and organ preservation, yet optimal expansion margins for rectal lesions remain unclear. This study aims to utilize high-resolution real-time imaging from MR-guided radiotherapy to characterize rectal motion uncertainties, optimize planning target volume (PTV) margins, and validate dosimetric outcomes, thereby enhancing the precision in rectal cancer dose escalation. A cohort of 22 advanced rectal cancer patients undergoing neoadjuvant radiotherapy on a 1.5T MR-Linac. Pre-treatment, intra-fraction, and post-treatment MR images were acquired for each fraction, with the rectum delineated according to the distance from the anus (upper, middle, lower). Inter- and intrafractional margins were determined by isotropic expansion of the segmented clinical target volume (CTV), with adequacy defined as covering 95% of the CTV volume in 90% of all fractions. Dosimetric validation was performed on 15 patients (10 internal, 5 external), comparing the segmental margins with literature-based uniform margins to assess target coverage and organs at risk (OARs) sparing. Rectal motion variability was segment-dependent, with the upper rectum requiring the largest margins and the lower rectum the smallest. Interfractional margins for the upper, middle, and lower rectum were 13 mm, 7 mm, and 6.5 mm, respectively. Intrafractional analysis revealed that treatment duration also influenced margin requirements, with 7 mm for the upper rectum and 3 mm for the middle and lower rectum sufficient for durations under 30 min; longer durations necessitated slightly larger margins. Dosimetric analysis confirmed that the segmented margin expansion strategy achieves adequate target coverage while substantially sparing OARs with dose reductions ranging from 7.4% to 44.5%. Segment-specific margin expansion suggest a potential improvement for rectal radiotherapy, ensuring target coverage while reducing OAR dose, thereby facilitating organ and function preservation. Not applicable.
Yang et al. (Tue,) studied this question.