Accuracy of dose-volume metric calculation for small-volume radiosurgery targets

Clinical impact of DVH uncertainties

Dose-volume histograms (DVH), along with dose and volume metrics, are central to radiotherapy planning. As such, errors have the potential to significantly impact the selection of appropriate treatment plans. Dose distributions that pass tests in one TPS may fail the same tests when transferred to another, even if using identical structures and dose grid information. This work shows the design and implementation of methods for assessing the accuracy of dose and volume computations performed by treatment planning systems (TPS), and other analytical tools. We demonstrate examples where differences in calculations between systems can change the assessment of a plan's clinical acceptability. Our work also provides a more detailed DVH analysis of single targets than earlier published studies. This is relevant for SRS plans and small structure dose assessments. Very small structures are a particular problem because of their coarse digital representation, and the impact of this is thoroughly examined. Reference DVH curves were derived mathematically, based on Gaussian dose distributions centered on spherical structures. The structures and dose distributions were generated synthetically, and imported into RayStation, MasterPlan, and ProKnow. Corresponding DVHs were analytically derived and taken as ground truth references, for comparison with the commercial DVH calculations. Two commonly used dose metrics PCI and MGI were used to determine the limit of calculation accuracy for small structures. In addition, to measure the DVH differences between a larger range of commercial DVH calculators, the D95 metric from a set of real clinical plans was compared across both the 3 DVH calculators under test, and across a further six TPSs from other hospitals. We show that even slight deviations between the results of DVH calculators can lead to plan check failures, and we illustrate this with the commonly used D95 planning metric. We present clinical data across eight planning systems that highlight instances where plan checks would pass in one software and fail in another due to DVH calculation differences. For the smallest volumes tested, errors of up to 20% were observed in the DVHs. RayStation was tested down to a 3 mm radius sphere (≈0.1 cc) and this showed close to 10% error, reducing to 1% for 10 mm radius (≈4.0 cc) and 0.1% for 20 mm radius (≈33 cc). In clinical plans, the variation in D95 was up to 9% for the smallest volumes, and typically around 2% in the range 0.5 cc-20 cc, and 1% in 20 cc-70 cc, falling to <0.1% for large volumes. Paddick Conformity Index (PCI) and Modified Gradient Index (MGI) are commonly used plan quality indicators for very small volumes. For volumes ≈0.1 cc we observed errors of up to 40% in PCI, and up to 75% in MGI. Our study extends the range of tested DVH calculators in published work, and shows their performance over a wider range of volume sizes. We provide quantitative evidence of the critical need to test the accuracy of DVH calculators in the TPS before clinical use. This work is particularly relevant for both stereotactic plan evaluation and for assessment of small volume doses in published dose constraint recommendations. We demonstrate that significant errors can occur in DVHs for volumes less than 1 cc, even if the volumes themselves are calculated accurately. Even for large structures, deviations between the outputs of DVH calculators can lead to indicated or reported plan check failures if they do not include appropriate tolerances. We urge caution in the use of DVH metrics for these very small volumes and recommend that appropriate DVH uncertainty tolerances are set in organ dose constraints when using them to evaluate clinical plans.

Dosimetric comparison of HyperArc and InCise MLC-based CyberKnife plans in treating single and multiple brain metastases

BACKGROUND AND PURPOSE

This study aimed to compare the dosimetric attributes of two multi-leaf collimator based techniques, HyperArc and Incise CyberKnife, in the treatment of brain metastases.

MATERIAL AND METHODS

17 cases of brain metastases were selected including 6 patients of single lesion and 11 patients of multiple lesions. Treatment plans of HyperArc and CyberKnife were designed in Eclipse 15.5 and Precision 1.0, respectively, and transferred to Velocity 3.2 for comparison.

RESULTS

HyperArc plans provided superior Conformity Index (0.91 ± 0.06 vs. 0.77 ± 0.07, p < 0.01) with reduced dose distribution in organs at risk (Dmax, p < 0.05) and lower normal tissue exposure (V4Gy-V20Gy, p < 0.05) in contrast to CyberKnife plans, although the Gradient Indexes were similar. CyberKnife plans showed higher Homogeneity Index (1.54 ± 0.17 vs. 1.39 ± 0.09, p < 0.05) and increased D2% and D50% in the target (p < 0.05). Additionally, HyperArc plans had significantly fewer Monitor Units (MUs) and beam-on time (p < 0.01).

CONCLUSION

HyperArc plans demonstrated superior performance compared with MLC-based CyberKnife plans in terms of conformity and the sparing of critical organs and normal tissues, although no significant difference in GI outcomes was noted. Conversely, CyberKnife plans achieved a higher target dose and HI. The study suggests that HyperArc is more efficient and particularly suitable for treating larger lesions in brain metastases.

Multicenter Multivendor Evaluation of Dose Volume Histogram Creation Consistencies for 8 Commercial Radiation Therapy Dosimetric Systems

PURPOSE

To evaluate dose volume histogram (DVH) construction differences across 8 major commercial treatment planning systems (TPS) and dose reporting systems for clinically treated plans of various anatomic sites and target sizes.

METHODS AND MATERIALS

Dose files from 10 selected clinically treated plans with a hypofractionation, stereotactic radiation therapy prescription or sharp dose gradients such as head and neck plans ranging from prescription doses of 18 Gy in 1 fraction to 70 Gy in 35 fractions, each calculated at 0.25 and 0.125 cm grid size, were created and anonymized in Eclipse TPS, and exported to 7 other major TPS (Pinnacle, RayStation, and Elements) and dose reporting systems (MIM, Mobius, ProKnow, and Velocity) systems for comparison. Dose-volume constraint points of clinical importance for each plan were collected from each evaluated system (D0.03 cc [Gy], volume, and the mean dose were used for structures without specified constraints). Each reported constraint type and structure volume was normalized to the value from Eclipse for a pairwise comparison. A Wilcoxon rank-sum test was used for statistical significance and a multivariable regression model was evaluated adjusting for plan, grid size, and distance to target center.

RESULTS

For all DVH points relative to Eclipse, all systems reported median values within 1.0% difference of each other; however, they were all different from Eclipse. Considering mean values, Pinnacle, RayStation, and Elements averaged at 1.038, 1.046, and 1.024, respectively, while MIM, Mobius, ProKnow, and Velocity reported 1.026, 1.050, 1.033, and 1.022, respectively relative to Eclipse. Smaller dose grid size improved agreement between the systems marginally without statistical significance. For structure volumes relative to Eclipse, larger differences are seen across all systems with a range in median values up to 3.0% difference and mean up to 10.1% difference.

CONCLUSIONS

Large variations were observed between all systems. Eclipse generally reported, at statistically significant levels, lower values than all other evaluated systems. The nonsignificant change resulting from lowering the dose grid resolution indicates that this resolution may be less important than other aspects of calculating DVH curves, such as the 3-dimensional modeling of the structure.

Aleatoric and epistemic uncertainty extraction of patient-specific deep learning-based dose predictions in LDR prostate brachytherapy

Objective.In brachytherapy, deep learning (DL) algorithms have shown the capability of predicting 3D dose volumes. The reliability and accuracy of such methodologies remain under scrutiny for prospective clinical applications. This study aims to establish fast DL-based predictive dose algorithms for low-dose rate (LDR) prostate brachytherapy and to evaluate their uncertainty and stability.Approach.Data from 200 prostate patients, treated with125I sources, was collected. The Monte Carlo (MC) ground truth dose volumes were calculated with TOPAS considering the interseed effects and an organ-based material assignment. Two 3D convolutional neural networks, UNet and ResUNet TSE, were trained using the patient geometry and the seed positions as the input data. The dataset was randomly split into training (150), validation (25) and test (25) sets. The aleatoric (associated with the input data) and epistemic (associated with the model) uncertainties of the DL models were assessed.Main results.For the full test set, with respect to the MC reference, the predicted prostateD90metric had mean differences of -0.64% and 0.08% for the UNet and ResUNet TSE models, respectively. In voxel-by-voxel comparisons, the average global dose difference ratio in the [-1%, 1%] range included 91.0% and 93.0% of voxels for the UNet and the ResUNet TSE, respectively. One forward pass or prediction took 4 ms for a 3D dose volume of 2.56 M voxels (128 × 160 × 128). The ResUNet TSE model closely encoded the well-known physics of the problem as seen in a set of uncertainty maps. The ResUNet TSE rectum D2cchad the largest uncertainty metric of 0.0042.Significance.The proposed DL models serve as rapid dose predictors that consider the patient anatomy and interseed attenuation effects. The derived uncertainty is interpretable, highlighting areas where DL models may struggle to provide accurate estimations. The uncertainty analysis offers a comprehensive evaluation tool for dose predictor model assessment.

A Monte Carlo dose recalculation pipeline for durable datasets: an I-125 LDR prostate brachytherapy use case

Monte Carlo (MC) dose datasets are valuable for large-scale dosimetric studies. This work aims to build and validate a DICOM-compliant automated MC dose recalculation pipeline with an application to the production of I-125 low dose-rate prostate brachytherapy MC datasets. Built as a self-contained application, the recalculation pipeline ingested clinical DICOM-RT studies, reproduced the treatment into the Monte Carlo simulation, and outputted a traceable and durable dose distribution in the DICOM dose format. MC simulations with TG43-equivalent conditions using both TOPAS andegs_brachyMC codes were compared to TG43 calculations to validate the pipeline. The consistency of the pipeline when generating TG186 simulations was measured by comparing simulations made with both MC codes. Finally,egs_brachysimulations were run on a 240-patient cohort to simulate a large-scale application of the pipeline. Compared to line source TG43 calculations, simulations with both MC codes had more than 90% of voxels with a global difference under ±1%. Differences of 2.1% and less were seen in dosimetric indices when comparing TG186 simulations from both MC codes. The large-scale comparison ofegs_brachysimulations with treatment planning system dose calculation seen the same dose overestimation of TG43 calculations showed in previous studies. The MC dose recalculation pipeline built and validated against TG43 calculations in this work efficiently produced durable MC dose datasets. Since the dataset could reproduce previous dosimetric studies within 15 h at a rate of 20 cases per 25 min, the pipeline is a promising tool for future large-scale dosimetric studies.

A comprehensive evaluation of advanced dose calculation algorithms for brain stereotactic radiosurgery

PURPOSE

Accurate dose calculation is important in both target and low dose normal tissue regions for brain stereotactic radiosurgery (SRS). In this study, we aim to evaluate the dosimetric accuracy of the two advanced dose calculation algorithms for brain SRS.

METHODS

Retrospective clinical case study and phantom study were performed. For the clinical study, 138 SRS patient plans (443 targets) were generated using BrainLab Elements Voxel Monte Carlo (VMC). To evaluate the dose calculation accuracy, the plans were exported into Eclipse and recalculated with Acuros XB (AXB) algorithm with identical beam parameters. The calculated dose at the target center (Dref), dose to 95% target volume (D95), and the average dose to target (Dmean) were compared. Also, the distance from the skull was analyzed. For the phantom study, a cylindrical phantom and a head phantom were used, and the delivered dose was measured by an ion chamber and EBT3 film, respectively, at various locations. The measurement was compared with the calculated doses from VMC and AXB.

RESULTS

In clinical cases, VMC dose calculations tended to be higher than AXB. It was found that the difference in Dref showed > 5% in some cases for smaller volumes < 0.3 cm3 . Dmean and D95 differences were also higher for small targets. No obvious trend was found between the dose difference and the distance from the skull. In phantom studies, VMC dose was also higher than AXB for smaller targets, and VMC showed better agreement with the measurements than AXB for both point dose and high dose spread.

CONCLUSION

The two advanced calculation algorithms were extensively compared. For brain SRS, AXB sometimes calculates a noticeable lower target dose for small targets than VMC, and VMC tends to have a slightly closer agreement with measurements than AXB.

The dosimetric impact of titanium implants in spinal SBRT using four commercial treatment planning algorithms

To evaluate the dosimetric impact of titanium implants in spine SBRT using four dose calculation algorithms. Twenty patients with titanium implants in the spine treated with SBRT without density override (DO) were selected. The clinical plan for each patient was created in Pinnacle and subsequently imported into Eclipse (AAA and AcurosXB) and Raystation (CC) for dose evaluation with and without DO to the titanium implant. We renormalized all plans such that 90% of the tumor volume received the prescription dose and subsequently evaluated the following dose metrics: (1) the maximum dose to 0.03 cc (Dmax), dose to 99% (D99%) and 90% (D90%) of the tumor volume; (2) Dmax and volumetric metrics of the spinal cord. For the same algorithm, plans with and without DO had similar dose distributions. Differences in Dmax, D99% and D90% of the tumor were on average <2% with slightly larger variations up to 5.58% in Dmax using AcurosXB. Dmax of the spinal cord for plans calculated with DO increased but the differences were clinically insignificant for all algorithms (mean: 0.36% ± 0.7%). Comparing to the clinical plans, the relative differences for all algorithms had an average of 1.73% (-10.36%-13.21%) for the tumor metrics and -0.93% (-9.87%-10.95%) for Dmax of the spinal cord. A few cases with small tumor and spinal cord volumes, dose differences of >10% in both D99% and Dmax of the tumor, and Dmax of the spinal cord were observed. For all algorithms, the presence of titanium implants in the spine for most patients had minimal impact on dose distributions with and without DO. For the same plan calculated with different algorithms, larger differences in volumetric metrics of >10% could be observed, impacted by dose gradient at the plan normalization volume, tumor volumes, plan complexity, and partial voxel volume interpolation.

Evaluation of a dedicated software for semi-automated VMAT planning of spine Stereotactic Body Radiotherapy (SBRT)

PURPOSE

To determine whether SBRT of spinal metastasis using a dedicated treatment planning system (TPS) and delivered with a gantry-based LINAC could provide plans of similar quality to the Cyberknife technology. Additional comparison was also done with other commercial TPS used for volumetric modulated arc therapy (VMAT) planning.

MATERIALS AND METHODS

Thirty Spine SBRT patients, previously treated in our institution with CyberKnife (Accuray, Sunnyvale) using Multiplan TPS, were replanned in VMAT with an dedicated TPS (Elements Spine SRS, Brainlab, Munich) and our clinical TPS (Monaco, Elekta LTD, Stockholm), using exactly the same arc geometry. The comparison was done by assessing differences in dose delivered to PTV, CTV and spinal cord, calculating modulation complexity scores (MCS) and performing quality control (QA) of the plans.

RESULTS

Regardless of the vertebra level, in general, no statistical difference was found in PTV coverage between all TPS. Conversely, PTV and CTV D_50% were found significantly higher for the dedicated TPS compared to others. In addition, the dedicated TPS also resulted in better gradient index (GI) than clinical VMAT TPS, whatever the vertebral level, and better GI than Cyberknife TPS for the thoracic level only. The D_2% to the spinal cord was generally significantly lower with the dedicated TPS compared with others. No significant difference was found in the MCS between both VMAT TPS. All QA were clinically acceptable.

CONCLUSION

The Elements Spine SRS TPS offers very effective and user-friendly semi-automated planning tools and is secure and promising for gantry-based LINAC spinal SBRT.

The effect of tumor shape irregularity on Gamma Knife treatment plan quality and treatment outcome: an analysis of 234 vestibular schwannomas

The primary aim of Gamma Knife (GK) radiosurgery is to deliver high-dose radiation precisely to a target while conforming to the target shape. In this study, the effects of tumor shape irregularity (TSI) on GK dose-plan quality and treatment outcomes were analyzed in 234 vestibular schwannomas. TSI was quantified using seven different metrics including volumetric index of sphericity (VioS). GK treatment plans were created on a single GK-Perfexion/ICON platform. The plan quality was measured using selectivity index (SI), gradient index (GI), Paddick's conformity index (PCI), and efficiency index (EI). Correlation and linear regression analyses were conducted between shape irregularity features and dose plan indices. Machine learning was employed to identify the shape feature that predicted dose plan quality most effectively. The treatment outcome analysis including tumor growth control and serviceable hearing preservation at 2 years, were conducted using Cox regression analyses. All TSI features correlated significantly with the dose plan indices (P < 0.0012). With increasing tumor volume, vestibular schwannomas became more spherical (P < 0.05) and the dose plan indices varied significantly between tumor volume subgroups (P < 0.001 and P < 0.01). VioS was the most effective predictor of GK indices (P < 0.001) and we obtained 89.36% accuracy (79.17% sensitivity and 100% specificity) for predicting PCI. Our results indicated that TSI had significant effects on the plan quality however did not adversely affect treatment outcomes.

Assessment of Dose-Volume Histogram Precision for Five Clinical Systems

PURPOSE

To investigate the dependency of dose-volume histogram behavior (DVH) and precision on underlying DICOM discretization using shapes and dose distributions with known analytical DVHs for five commercial DVH calculators.

METHODS

DVHs and summary metrics were extracted from all five systems using synthetic DICOM cone and cylinder objects for which the true volume and DVH curves were known. Trends in the curves and metrics were explored by varying the underlying voxelization of the CT image, structure set, and dose grid as well by varying the geometry of the structure and direction of a linear dose gradient. Using synthetic structures allowed for comparison with ground-truth DVH curves to assess their accuracy while an algorithm was additionally developed to assess the precision of each system. The precision was calculated with a novel algorithm that treats any "stair step" behavior in a DVH curve as an uncertainty band and calculates the width, characterized as a percent difference, of the band for various DVH metrics. The underlying voxelization was additionally changed and DVHs were extracted for two clinical examples. The details of how each system calculated DVHs were also investigated and tendencies in the calculated curves, metrics, and precision were related to choices made in the calculation methodology.

RESULTS

Calculation methodology differences that had a noticeable impact on the DVH curves and summary metrics include supersampling beyond the input grids and interpretation of the superior and inferior ends of the structures. Amongst the systems studied, the median precision ranged from 0.902% to 3.22%, and interquartile ranges varied from 1.09% to 3.91%.

CONCLUSIONS

Commercial dose-evaluation solutions can calculate different DVH curves, structure volume measures, and dose statistics for the same input data due to differences in their calculation methodologies. This study highlights the importance of understanding and investigating the DVH calculation when considering a new clinical system and when using more than one system for data transfer. This article is protected by copyright. All rights reserved.

Treatment planning system commissioning of the first clinical biology‐guided radiotherapy machine

Purpose

Methods

Results

Conclusions

Evaluation of two automated treatment planning techniques for multiple brain metastases using a single isocenter

Two automated treatment planning techniques were evaluated for multiple brain metastases using a single isocenter. One technique is knowledge-based planning (KBP) using a stereotactic radiosurgery (SRS) model in Eclipse treatment planning system (TPS); and the other is the Multiple Brain Mets (MBM) SRS technique in Brainlab Elements TPS. Eighteen plans each with 3-10 lesions were used for the study. Plan evaluation metrics included the planning target volume (PTV) coverage, conformity index (CI), total monitor units (MUs), plan optimization time, brain V_12 Gy, V_8 Gy, and V_5 Gy. Both the KBP and MBM planning techniques produced comparable plans to the manually generated clinical plans in terms of PTV coverage and CI. For irregularly shaped lesions, the KBP plans provided more conformal dose distribution to the PTV than the MBM plans. The KBP plans took significantly longer time to plan but have fewer MUs than the MBM plans. The MBM plans spared normal brain tissues better than the KBP plans in terms of V_5 Gy.