A Statistical Approach to Quantification and Visualization of Setup and Range Uncertainties for Proton Plan Verification
P Park1*, J Cheung1, X Zhu1, W Liu1, N Sahoo1, L Court1, R Mohan1, H Li1, L Dong2, (1) The University of Texas MD Anderson, Houston, TX, (2) Scripps Proton Therapy Center, San Diego, CATH-A-116-1 Thursday 8:00AM - 9:55AM Room: 116
Proton therapy is more sensitive to both setup and beam range uncertainties than the conventional photon therapy due to the finite range of protons. For the purpose of evaluation of plan robustness in the face of such uncertainties, we developed a method to quantify and visualize the uncertainties in proton plans using comprehensive statistical analysis. The application of the proposed method is shown for selected patients in various clinical sites.
Clinically approved 20 lung, 10 prostate, and a brain cases were retrospectively analyzed. For a selected volume of interest, probability density function (PDF) describing variation in point dose value for every tissue element under setup and range uncertainties was computed based on over 600 dose calculations per plan. The expected dose-volume histogram (E-DVH) was calculated by weighting the probability of each tissue element receiving a specified dose level. Similarly, standard deviation of the E-DVH was calculated (SD-DVH). Furthermore for each volume element, their probability of receiving greater dose than the dose constraint was visualized as color-wash, mapped onto the patients CT images for spatial visualization and assessment.
The E-DVH value of target coverage under the uncertainties was consistently lower than that of the nominal value, with a mean difference of -1.1% (-0.9% for breath-hold), -0.3%, and -2.2% for lung, prostate, and a brain cases respectively. The organs that were most sensitive to the uncertainties were esophagus and spinal cord for lung, rectum for prostate, and brain stem for brain cancer.
A clinically feasibly plan robustness analysis tool based on statistical simulation has been developed for proton therapy plans. The passively scattered beam lung plans and scanning beam prostate plans appears to be robust in terms of target coverage. However, structures that are small in volume or located near the target area showed greater sensitivity to uncertainties.
Funding Support, Disclosures, and Conflict of Interest: This project is supported by grant P01CA021239 from the National Cancer Institute.