Purpose: To investigate different MRI-compatible shield designs for dynamically-rotating intensity modulated brachytherapy (IMBT) of rectal cancer and compare the results to conventional HDR brachytherapy (HDR-BT) with and without static shielding.
Methods: Three rigid tungsten shields with different emission windows (2 mm, 90°, 180°) and three flexible shields with 180° emission windows were designed for rectal IMBT. All shields had a 2 mm central source channel and a 15 mm diameter. Dosimetric properties of the IMBT shields were explored in water using RapidBrachyMCTPS, a Monte Carlo-based treatment planning system, and a generic Ir-192 source. Additionally, to calculate the absorbed dose to the tumor and surrounding healthy tissue, data from one rectal cancer patient treated with three fractions of HDR-BT was used retrospectively. Dwell positions were created along the source channel with a 5 mm step size and 15° increment shield rotation. Simulations were repeated using the conventional applicator with and without static shielding for comparison.
Results: Highly anisotropic dose distributions were achieved with all IMBT shields. For the same CTV coverage, the rigid 180° shield resulted in better sparing of the contralateral healthy rectal wall, superior dose spill region, rectum, bladder, femur and pelvis compared to the conventional applicator with and without static shielding. Only the inferior D2cc was greater by up to 14% compared to the conventional applicator with static shielding. The mean treatment time exceeded 20 min for the rigid 2 mm and 90° shields, which is not clinically acceptable. All simulations achieved an uncertainty of less than 1% on the absorbed dose for values within the 100% isodose line.
Conclusion: The dosimetric impact of three rigid and three flexible shield designs for rectal IMBT was quantitatively evaluated. The next step is to perform a larger retrospective study with the rigid 180° shield.