Program Information
Dosimetric Analysis of Patient-Specific Planning Target Volumes Using Population Statistical Modeling of Interfractional Prostate Motion
D Vile1*, G Christensen2 , J Ford1 , N Mukhopadhyay1 , J Williamson1 , (1) Virginia Commonwealth University, Richmond, VA, (2) University of Iowa, Iowa City, IA
Presentations
TH-EF-BRD-9 (Thursday, July 16, 2015) 1:00 PM - 2:50 PM Room: Ballroom D
Purpose: To use a population-based statistical motion model to create a patient-specific planning target volume (PTV) for patients undergoing radiotherapy for definitive treatment of low-risk prostate cancer and to compare the dosimetric impact of this PTV against current clinically used PTVs.
Methods: From mapping 10-13 fractional CT images to the planning image, systematic and random deformations were calculated for 19 prostate patients. These deformations were mapped to a reference image using inter-patient deformable image registration (DIR). In the reference image coordinate system, principal component analysis (PCA) was used to create 3D statistical motion models of the systematic and random displacements. Using these models to sample synthetic displacements and mapping them back to a patient’s planning coordinate system, organ occupancy maps were created for the prostate. The organ occupancy map was thresholded at the 95% coverage level to create a PCA-based PTV. For bony aligned and prostate centroid aligned patient setups, a virtual clinical trial was conducted to determine the dosimetric differences between a PCA-based PTV and a conventional PTV (using a van Herk margin for bony alignment and 5 mm fixed [3 mm posterior] for centroid alignment). Dose volume histogram (DVH) metrics were used in the analysis.
Results: All PTVs provided prescription dose coverage to the prostate. For bony aligned setup, the PCA-based PTV significantly (p<0.05) reduced D₃₀, D₂₀, and D₅ to bladder and D₅₀ to rectum, while increasing rectal D₂₀ and D₅. For the centroid aligned setup, the PCA PTV significantly reduced all bladder DVH metrics and trended to lower rectal DVH metrics.
Conclusion: Statistical motion modeling provides a systematic method for creating PTVs. Plans created using these PTVs provide adequate prescription dose coverage of the prostate while lowering the bladder dose. PCA-based PTVs tended to increase rectal dose for bony-aligned patients and lower it for centroid aligned patients.
Contact Email: