Author: Emily Helen Hayes, Chihray Liu 👨🔬
Affiliation: University of Florida 🌍
Purpose: To evaluate dosimetric discrepancies in bladder dose calculations among rigid, deformed, and manual contouring methods in prostate cancer patients and assess dose variations resulting from bladder volume changes.
Methods: Five prostate cancer patients treated on a MR-Linac with 60Gy over 20 fractions were retrospectively analyzed. Prior to treatment, patients were instructed to follow fasting protocols, though variations were observed in daily MRI scans. These scans were fused with an initial CT to overlay the patient’s contoured anatomy and adjust for daily shifts. Rigid contours were unchanged from the CT’s original delineations, deformed contours were adjusted automatically by treatment planning software to align with the MRI bladder outline, and manual contours were drawn by hand on every 3mm slice. Bladder volumes were classified as low, intermediate, or high based on the mean and standard deviation of the individual patients’ volumes. Dose calculations were evaluated at bladder volumes 0.03cc, 5.0cc, 10.00cc and 20.00cc surrounding the PTV. Approved dosimetric criteria were 63Gy, 60Gy, 55Gy, and 50Gy, respectively. A DVH statistic set was calculated for each combination of contouring method, bladder volume, and evaluated volume. The differences were calculated, averaged, and divided by the dosimetric criteria to determine percent deviation.
Results: Dosimetric differences from manual contouring ranged from 0.471 Gy (0.748%) to 2.064 Gy (4.128%) for rigid structures and between 0.354 Gy (0.562%) to 1.064 Gy (3.208%) for the deformed structures. Low-volume bladders experienced the largest deviations between manual and automatic contouring methods.
Conclusion: This study demonstrates that manual and automated contouring methods yield comparable dose calculations, supporting the feasibility of automated approaches to significantly reduce daily treatment planning time. These findings enhance workflow efficiency and patient comfort in MR-guided radiotherapy. Future work will extend this methodology to other variable organs, such as the rectum.