Comparing Dose Summation Techniques for External Beam and LDR Brachytherapy for Combined Modality Prostate Cancer Treatment
A Cooney1,2*, S To1,2, D Guest2, A Riegel1,2,3, (1) Department of Radiation Medicine, Northwell Health, New Hyde Park, NY (2) Hofstra University, Hempstead, NY, (3) Zucker School of Medicine at Hofstra/Northwell, Hempstead, NY
Presentations
PO-GePV-M-142 (Sunday, 7/25/2021) [Eastern Time (GMT-4)]
Purpose: To compare composite biological effective dose (BED) of prostate cancer patients from external beam radiation therapy (EBRT) and low dose rate (LDR) brachytherapy contributions through image-based rigid registration, fiducial-based rigid registration, b-spline deformable image registration (DIR), and dose volume histogram (DHV) parameter summing techniques.
Methods: Ninety one (91) patients received intensity-modulated EBRT before ¹⁰³Pd-based LDR. Post-implant LDR plan CTs were registered to EBRT planning CTs via image-based rigid registration, fiducial-based rigid registration, and DIR. EBRT and LDR physical doses were converted to BED and summed via the 3 image registration techniques and scalar summation of select DVH parameters without registration. Results were compared to summed dose of 44 “control” patients whose post-implant and EBRT planning CTs were inherently registered via DICOM because they received LDR before EBRT. Prostate, urethra, bladder, and rectum dosimetric parameters were compared for statistically significant differences using ANOVA (α=0.05).
Results: Among the three image registration techniques, prostate D90 was higher for fiducial-based rigid registration (122.6±19.6 GyBED) compared with image-based rigid and DIR (114.6±18.4 GyBED, 116.1±16.5 GyBED respectively), and all three were lower than the control group (130.1±19.7 GyBED). Scalar summation of prostate D90 was significantly higher than all 3 image registration techniques and the control group (168.3±17.2 GyBED), potentially due to interobserver variation in EBRT and LDR prostate contour. Composite urethra D10, rectum D2cc, and rectum D0.1cc from image-based rigid registration were significantly higher than the other registration techniques (3-7% for urethra, 16-20% for rectum D2cc, 26-30% for rectum D0.1cc.)
Conclusion: Fiducial-based rigid registration and deformable registration yielded combined BED distributions that were closest (within 5-10%) to BED distributions from inherently-registered treatment plans. Scalar summation of DVH parameters may overestimate combined dosimetry, particularly with discrepancies in contour delineation between external beam and brachytherapy treatment plans.
ePosters
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email