摘要

Purpose: To develop a probabilistic treatment planning (PTP) method which is robust to systematic patient setup errors and to compare PTP plans with plans generated using a planning target volume (PTV) margin optimized to give the same target coverage probability as the PTP plan. %26lt;br%26gt;Methods: Plans adhering to the RTOG-0126 protocol are developed for 28 prostate patients using PTP and margin-based planning. For PTP, an objective function that simultaneously considers multiple possible patient positions is developed. PTP plans are optimized using clinical target volume (CTV) structures and organ at risk (OAR) structures. The desired CTV coverage probability is 95%. Plans that cannot achieve a 95% CTV coverage probability are re-optimized with a desired CTV coverage probability reduced by 5% until the desired CTV coverage probability is achieved. Margin-based plans are created which achieve the same CTV coverage probability as the PTP plans by iterative adjustment of the CTV-to-PTV margin. Postoptimization, probabilistic dose-volume coverage metrics are used to compare the plans. %26lt;br%26gt;Results: For equivalent target coverage probability, PTP plans significantly reduce coverage probability for rectum objectives (-17% for D-35 %26lt; 65 Gy, p = 0.0010; -23% for D-25 %26lt; 70 Gy, p %26lt; 0.0001; and -27% for D-15 %26lt; 75 Gy, p %26lt; 0.0001). Physician assessment indicates PTP plans are entirely preferred 71% of the time while margin-based plans are entirely preferred 7% of the time. %26lt;br%26gt;Conclusions: For plans having the same target coverage probability, PTP has potential to reduce rectal doses while maintaining CTV coverage probability. In blind comparisons, physicians prefer PTP plans over optimized margin plans. (Work supported by NIH P01CA116602 and T32CA113277.).

  • 出版日期2012-2