Geometrical tracking accuracy and appropriate PTV margins for robotic radiosurgery of liver lesions by SBRT

Dosimetric impact of rotational errors in trigeminal neuralgia radiosurgery using CyberKnife

PURPOSE

Trigeminal neuralgia (TN) can be treated on the CyberKnife system using two different treatment delivery paths: the general-purpose full path corrects small rotations, while the dedicated trigeminal path improves dose fall-off but does not allow rotational corrections. The study evaluates the impact of uncorrected rotations on brainstem dose and the length of CN5 (denoted as Leff) covered by the prescription dose.

METHODS AND MATERIALS

A proposed model estimates the delivered dose considering translational and rotational delivery errors for TN treatments on the CyberKnife system. The model is validated using radiochromic film measurements with and without rotational setup error for both paths. Leff and the brainstem dose is retrospectively assessed for 24 cases planned using the trigeminal path. For 15 cases, plans generated using both paths are compared for the target coverage and toxicity to the brainstem.

RESULTS

In experimental validations, measured and estimated doses agree at 1%/1 mm level. For 24 cases, the treated Leff is 5.3 ± 1.7 mm, reduced from 5.9 ± 1.8 mm in the planned dose. Constraints for the brainstem are met in 23 cases for the treated dose but require frequent treatment interruption to maintain rotational corrections <0.5° using the trigeminal path. The treated length of CN5, and plan quality metrics are similar for the two paths, favoring the full path where rotations are corrected.

CONCLUSIONS

We validated an analytical model that can provide patient-specific tolerances on rotations to meet plan objectives. Treatment using the full path can reduce treatment time and allow for rotational corrections.

Preoperative single fraction breast radiotherapy: Intra-fraction geometric uncertainties and dosimetric implications

PURPOSE

A preoperative breast robotic radiosurgery trial was concluded in our centre. Purposes of the present study were to evaluate retrospectively over the enrolled patients: i) respiratory patterns ii) tracking uncertainties iii) necessity of respiratory compensation iv) tracking errors dosimetric effects.

METHODS

22 patients were treated in 21 Gy single fraction using CyberKnife (CK) respiratory modelling and tracking (SynchronyResp) and data extracted from log-files. Respiratory motion and baseline drifts (BD) were analyzed. SynchronyResp uncertainties were computed and compared with errors simulated for CK fiducial tracking without respiratory compensation. Plans were perturbed by tracking errors and perturbed doses calculated on the planning CT scan in order to simulate the dosimetric consequences of intra-fraction errors.

RESULTS

After BD correction, respiratory amplitudes were below 5.5 mm except one value of 8 mm. 50% of patients showed BD above 3 mm. Standard deviations of SynchronyResp errors remained within 2.1 mm. Standard deviations of tracking errors without respiratory compensation were comparable and below 2.5 mm. Using a 3 mm PTV margin, perturbed CTV coverage was below 95% (93.7%) just for one patient. The latter case presented a large CTV-Skin interface. Perturbed OAR doses were always judged clinically acceptable.

CONCLUSION

Intra-fraction geometric uncertainties and their effects were quantified for breast neoadjuvant CK treatments. Data indicated that in the majority of cases respiratory compensation may be disabled without increasing uncertainties and reducing treatment time, provided that fiducial intra-fraction tracking is performed to account for BD. Dosimetric effects are mostly not clinically relevant.

Comprehensive Treatment Uncertainty Analysis and PTV Margin Estimation for Fiducial Tracking in Robotic Liver Stereotactic Body Radiation Therapy

OBJECTIVE

This study aims to quantify the uncertainties of CyberKnife Synchrony fiducial tracking for liver stereotactic body radiation therapy (SBRT) cases, and evaluate the required planning target volume (PTV) margins.

METHODS

A total of 11 liver tumor patients with a total of 57 fractions, who underwent SBRT with synchronous fiducial tracking, were enrolled for the present study. The correlation/prediction model error, geometric error, and beam targeting error were quantified to determine the patient-level and fraction-level individual composite treatment uncertainties. The composite uncertainties and multiple margin recipes were compared for scenarios with and without rotation correction during treatment.

RESULTS

The correlation model error-related uncertainty was 4.3±1.8, 1.4±0.5 and 1.8±0.7 mm in the superior-inferior (SI), left-right, and anterior-posterior directions, respectively. These were the primary contributors among all uncertainty sources. The geometric error significantly increased for treatments without rotation correction. The fraction-level composite uncertainties had a long tail distribution. Furthermore, the generally used 5-mm isotropic margin covered all uncertainties in the left-right and anterior-posterior directions, and only 75% of uncertainties in the SI direction. In order to cover 90% of uncertainties in the SI direction, an 8-mm margin would be needed. For scenarios without rotation correction, additional safety margins should be added, especially in the superior-inferior and anterior-posterior directions.

CONCLUSION

The present study revealed that the correlation model error contributes to most of the uncertainties in the results. Most patients/fractions can be covered by a 5-mm margin. Patients with large treatment uncertainties might need a patient-specific margin.

A dose perturbation tool for robotic radiosurgery: Experimental validation and application to liver lesions

BACKGROUND

An analytical tool is empirically validated and used to assess the delivered dose to liver lesions accounting for different types of errors in robotic radiosurgery treatment.

MATERIAL AND METHODS

A tool is proposed to estimate the target doses taking into account the translation, rotation, and deformation of a target. Translational errors are modeled as a spatial convolution of the planned dose with a probability distribution function derived from treatment data. Rotations are modeled by rotating the target volume about the imaging isocenter. Target deformation is simulated as an isotropic target expansion or contraction based on changes in inter-fiducial spacing. The estimated dose is validated using radiochromic film measurements in nine experimental conditions, including in-phase and out-of-phase internal-and-external breathing motion patterns, with and without uncorrectable rotations, and for homogenous and heterogeneous phantoms. The measured dose is compared to the perturbed and planned doses using gamma analyses. This proposed tool is applied to assess the dose coverage for liver treatments using D99/Rx where D99 and Rx are the minimum target and prescription doses, respectively. These metrics are used to evaluate plan robustness to different magnitudes of rotational errors. Case studies are presented to illustrate how to improve plan robustness against delivery errors.

RESULTS

In the experimental validations, measured dose agrees with the estimated dose at the 2%/2 mm level. When accounting for translational and rotational tracking residual errors using this tool, approximately one-fifth of targets are considered underdosed (D99/Rx < 1.0). If target expansion or contraction is modeled, approximately one-third of targets are underdosed. The dose coverage can be improved if treatments are planned following proposed guidelines.

CONCLUSION

The dose perturbation model can be used to assess dose delivery accuracy and investigate plan robustness to different types of errors.

Quantification of Intrafraction and Interfraction Tumor Motion Amplitude and Prediction Error for Different Liver Tumor Trajectories in Cyberknife Synchrony Tracking

PURPOSE

To research the fiducial-based, real-time tracking intrafraction (during the fraction [intra-]) and interfraction (between fractions [inter-]) tumor respiration amplitude, motion trajectory, and prediction error and quantify their relationships for different types of motion trajectories during Cyberknife-based stereotactic ablation radiotherapy.

METHODS AND MATERIALS

Twelve patients with liver tumors were treated using a Cyberknife system, and 58 fractions were involved in this study. Real-time target motion tracking data were extracted and transformed from the robot coordinate system into the patient coordinate system by the rotation matrix. Only the time sessions of the beam on were studied according to the data information generated from the Cyberknife motion tracking system. The motion correlation model between the external marker signal and internal fiducial position was built to present the type of motion trajectory.

RESULTS

Using the correlation model as a function of external marker signal and internal fiducial position, we knew 4 motion trajectories mainly existed for liver cancer patients as follows: perfect linearity (group I), simple linearity (group II), hysteresis (group III), and area respiratory (group IV) patterns. More than half of the patients had a linear breathing trajectory. Analyzing all patients together, the intra-amplitudes were slightly less than those of the inter-amplitudes. The amplitude from large to small was in the superior-inferior, left-right and anterior-posterior directions, regardless of inter- and intra-amplitudes. Then, patients with a larger peak-to-peak have a larger standard deviation of amplitude and a larger amplitude in all fractions/sessions. The prediction errors of the linear motion trajectory were generally less than 1 mm. The prediction errors of the regular hysteresis breathing model were smaller than those of the irregular hysteresis model. Scattered breathing would result in a larger tracking error, such as the area respiratory trajectory. It was logical that prediction errors were larger for patients who showed much variation in their breathing amplitude.

CONCLUSIONS

This paper showed that the liver motion trajectory model included perfect linearity, sample linearity, hysteresis, and area. The linear motion trajectory presented the minimum tracking error and the best stability, and the hysteresis and area trajectory were the worst. Therefore, breathing management, including respiration training, control, and evaluation of motion trajectory in all directions, was significantly necessary during liver SABR treatment.

Patient-specific PTV margins for liver stereotactic body radiation therapy determined using support vector classification with an early warning system for margin adaptation

PURPOSE

An adaptive planning target volume (PTV) margin strategy incorporating a volumetric tracking error assessment after each fraction is proposed for robotic stereotactic body radiation therapy (SBRT) liver treatments.

METHODS AND MATERIALS

A supervised machine learning algorithm employing retrospective data, which emulates a dry-run session prior to planning, is used to investigate if motion tracking errors are <2 mm, and consequently, planning target volume (PTV) margins can be reduced. A fraction of data collected during the beginning of a treatment course emulates a dry-run session (mock) before planning. Twenty features are calculated using mock data and used for support vector classification (SVC). A treatment course is labeled as Class 1 if the maximum root-mean-square radial tracking error for all remaining fractions is below 2 mm, or Class 2 otherwise. We evaluate the classification using fivefold cross-validation, leave-one-out cross-validation, 500 repeated random subsampling cross-validation, and the receiver operating characteristic (ROC) metric. The classification is independently cross-validated on a cohort of 48 treatment plans for other anatomical sites. A per fraction assessment of volumetric tracking errors is performed for the standard 5 mm PTV margin (PTV_std ) for courses predicted as Class 2; or for a margin reduced by 2 mm (PTV_std-2mm ) for those predicted as Class 1. We perturb the gross tumor volume (GTV) by the tracking errors for each x-ray image acquisition and calculate the fractional GTV voxel occupancy probability (P_i ) inside the PTV for each treatment fraction i. For treatment courses classified as Class 1, an early warning system flags treatment courses having any P_i  < 0.99, and the subsequent treatments are proposed to be replanned using PTV_std .

RESULTS

The classification accuracies are 0.84 ± 0.06 using fivefold cross-validation, and 0.77 when validated using an independent testing set (other anatomical sites). Eighty percent of treatment courses are correctly classified using leave-one-out cross-validation. The sensitivity, precision, specificity, F1 score, and accuracy are 0.81 ± 0.09, 0.85 ± 0.08, 0.80 ± 0.11, 0.83 ± 0.06, and 0.80 ± 0.07, respectively, using 500 repeated random subsampling cross-validation. The area under the curve for the ROC metric is 0.87 ± 0.05. The four most important features for classification are related to standard deviations of motion tracking errors, the linearity between the target location and external LED marker positions, and marker radial motion amplitudes. Eleven of 64 cases predicted to be of Class 1 have 0.96 < P_i  < 0.99 for each treatment fraction, and require replanning using PTV_std . In comparison, the PTV_std always covers the perturbed GTVs with P_i  > 0.99 for all patients.

CONCLUSIONS

Support vector classification is proposed for the classification of different motion tracking errors for patient courses based on a mock session before planning for SBRT liver treatments. It is feasible to implement patient-specific PTV margins in the clinic, assisted with an early warning system to flag treatment courses that require replanning using larger PTV margins in an adaptive treatment strategy.