SU-E-T-625: Robustness Evaluation and Robust Optimization of IMPT Plans Based on Per-Voxel Standard Deviation of Dose Distributions.

Proton dose distributions, IMPT in particular, are highly sensitive to setup and range uncertainties. We report a novel method, based on per-voxel standard deviation (SD) of dose distributions, to evaluate the robustness of proton plans and to robustly optimize IMPT plans to render them less sensitive to uncertainties. For each optimization iteration, nine dose distributions are computed - the nominal one, and one each for ± setup uncertainties along x, y and z axes and for ± range uncertainty. SD of dose in each voxel is used to create SD-volume histogram (SVH) for each structure. SVH may be considered a quantitative representation of the robustness of the dose distribution. For optimization, the desired robustness may be specified in terms of an SD-volume (SV) constraint on the CTV and incorporated as a term in the objective function. Results of optimization with and without this constraint were compared in terms of plan optimality and robustness using the so called'worst case' dose distributions; which are obtained by assigning the lowest among the nine doses to each voxel in the clinical target volume (CTV) and the highest to normal tissue voxels outside the CTV. The SVH curve and the area under it for each structure were used as quantitative measures of robustness. Penalty parameter of SV constraint may be varied to control the tradeoff between robustness and plan optimality. We applied these methods to one case each of H&N and lung. In both cases, we found that imposing SV constraint improved plan robustness but at the cost of normal tissue sparing. SVH-based optimization and evaluation is an effective tool for robustness evaluation and robust optimization of IMPT plans. Studies need to be conducted to test the methods for larger cohorts of patients and for other sites. This research is supported by National Cancer Institute (NCI) grant P01CA021239, the University Cancer Foundation via the Institutional Research Grant program at the University of Texas MD Anderson Cancer Center, and MD Anderson’s cancer center support grant CA016672.

Liu W ，Mohan R 《-》

SU-E-T-565: Effectiveness of Robust Optimization in Head and Neck IMPT Dose Distributions.

The effectiveness of IMPT may be significantly diminished by range and patient setup uncertainties. The purpose of this presentation is to evaluate the ability of robust optimization methods to desensitize H&N IMPT plans to uncertainties and their impact on plan optimality. We use a robust optimization method, in which the objective function value for a given iteration is computed using the 'worst case' dose distribution. The conventionally optimized PTV-based IMPT plans and robustly optimized plans were generated for 14 head and neck cancer cases. The dose standard deviation was calculated for every voxel and used to compute 'standard-deviation volume histograms' (SVHs). The area under SVH curves was used to quantify the plan robustness. In addition, D1cc doses for spinal cord and brainstem, mean doses Dmean for oral cavity and parotids, and D1% doses for other organs were used to assess plan optimality. D5% and D95% doses are used to assess target dose coverage and homogeneity. The plan optimality and robustness are then compared statistically by the pair t-tests using SPSS 19.0 software. Compared with PTV-based optimization, robust optimization provides significantly more robust dose distribution for both targets and organs without sacrificing, and possibly even improving, the sparing of normal tissues. In addition, our robust optimization method also leads to more homogeneous dose distribution in targets with better prescription dose coverage. Improvements are statistically significant with p-value smaller than 0.05 for almost all end points compared. Robust optimization results in patient-specific, optimizer-determined, and effectively reduced margins compared to a predefined and fixed margin used in the PTV approach. The optimizer can find a desired beamlet weight solution from the degenerate solution space so that the dose distribution follows the changes in anatomical geometry and is minimally perturbed by uncertainties. Our results demonstrate the importance of robust optimization. This research is supported by National Cancer Institute (NCI) grant P01CA021239, the University Cancer Foundation via the Institutional Research Grant program at the University of Texas MD Anderson Cancer Center, and MD Anderson’s cancer center support grant CA0 16672.

Liu W ，Frank S ，Li X ，Li Y ，Zhang X ，Zhu X ，Mohan R ... -  《-》

SU-E-T-624: Comparison of PTV+PRV-Based Optimization and Robust Optimization in Intensity-Modulated Proton Therapy.

Robust optimization leads to IMPT plans that are more robust than and superior in optimality compared to PTV-based optimized plans. Robust optimization incorporates setup and range uncertainties, which implicitly adds margins to targets and organs-at-risk (OARs); whereas PTV-based optimization only considers setup uncertainties and adds margins only to targets in practice. The purpose of this work is to determine if the superiority of robustly optimized plans is due to not assigning margins to OARs during PTV-based optimization. Plan robustness and optimality of the PTV plus Planning organs-at-Risk Volume (PRVs)-based plans and robustly optimized plans were compared for 5 head and neck cancer cases and one rhabdomyosarcoma case. The PRVs were generated by expansion from OARs by 3 mm. 9 different dose distributions were computed - one each for ± setup uncertainties along three spatial directions, for ± range uncertainty, and the nominal dose distribution. The worst-case dose distribution was obtained by assigning the lowest dose among the 9 doses to each voxel in the target and the highest dose to each voxel outside the target. The DVHs from the worst-case dose were used to assess the plan optimality and robustness. D1cc doses for spinal cord and brainstem, mean doses Dmean for oral cavity and parotids, and D1% doses for other organs were also used to assess plan optimality. D5% and D95% doses are used to assess target dose coverage and homogeneity. For H&N cases, PTV+PRV-based optimization was inferior to robust optimization. However, PTV+PRV-based optimization yielded plans that spared OARs better than PTV-based optimization, although the target dose robustness and homogeneity were comparable to the PTV-based optimization. The same conclusions are also valid in the rhabdomyosarcoma case. We find that the PTV+PRV method can partly improve plan optimality, but it is still inferior to robust optimization method. This research is supported by National Cancer Institute (NCI) grant P01CA021239, the University Cancer Foundation via the Institutional Research Grant program at the University of Texas MD Anderson Cancer Center, and MD Anderson’s cancer center support grant CA016672.

Liu W ，Zhu X ，Li X ，Li Y ，Zhang X ，Frank S ，Mohan R ... -  《-》

Robust optimization of intensity modulated proton therapy.

Liu W ，Zhang X ，Li Y ，Mohan R ... -  《-》

A novel and individualized robust optimization method using normalized dose interval volume constraints (NDIVC) for intensity-modulated proton radiotherapy.

Intensity-modulated proton therapy (IMPT) is known to be sensitive to patient setup and range uncertainty issues. Multiple robust optimization methods have been developed to mitigate the impact of these uncertainties. Here, we propose a new robust optimization method, which provides an alternative way of robust optimization in IMPT, and is clinically practical, which will enable users to control the balance between nominal plan quality and plan robustness in a user-defined fashion. We calculated nine individual dose distributions which corresponded to one nominal and eight extreme scenarios caused by patient setup and proton beam's range uncertainties. For each voxel, the normalized dose interval (NDI) is defined as the full dose range variation divided by the maximum dose in all uncertainty scenarios (NDI = [max - min dose]/max dose), which was then used to calculate the normalized dose interval volume histogram (NDIVH) curves. The areas under the NDIVH curves were used to quantify plan robustness. A normalized dose interval volume constraint (NDIVC) applied to the target was incorporated to specify the desired robustness which was user-defined. Users could then explore the trade-off between nominal plan quality and plan robustness by adjusting the position of the NDIVCs on the NDIVH curves freely. We benchmarked our method using one lung, five head and neck (H&N), and three prostate cases by comparing our results to those derived using the voxel-wise worst-case robust optimization. Using the benchmark cases, our new method achieved quality IMPT plans comparable to those derived from the voxel-wise worst-case robust optimization for both nominal plan quality and plan robustness in general; even more conformal and more homogeneous target dose distributions in some cases, if proper NDIVCs were applied. The AUC under NDIVH, as a precise quantitative index of plan robustness, was consistent with DVH bandwidths. Additionally, we demonstrated the feasibility of adjusting the position of NDIVCs in the NDIVH curves which allowed users to explore the trade-off between nominal plan quality and plan robustness. The NDIVH-based robust optimization method provided a novel and individualized way of robust optimization in IMPT, and enables users to adjust the balance between nominal plan quality and plan robustness in a user-defined fashion. This method is applicable for continued improvement and developing the next generation of IMPT planning algorithms in the future.

Shan J ，Sio TT ，Liu C ，Schild SE ，Bues M ，Liu W ... -  《-》