The transition from frame-based brain stereotactic radiosurgery (SRS) to frameless delivery is supported by realtime intrafraction monitoring to ensure accurate delivery. The purpose of this study is to characterize these real-time motion traces in a large cohort of patients treated with frameless gated brain SRS and to develop patient-specific predictions of tolerance violations. Methods and Materials: SRS patients treated on the Gamma Knife Icon were immobilized using a device-specific thermoplastic head mask. A motion marker was fixed to the patient's nose, with gating and cone beam computed tomography (CBCT)-based corrections to the treatment at excursions from baseline exceeding 1.5 mm. The traces of 1446 fractions were analyzed according to magnitude (932 unique treatment plans for 462 unique individual patients), directional distribution of displacement, and stability. A neural network model was developed to predict interruptions based on a subset of trace data. Results: The average displacement of the marker in the first fraction of all patients was 0.62 AE 0.25 mm with beam CBCT corrections, which would otherwise be modeled at 0.96 AE 0.96 mm without intrafraction motion correction (P < .0001). Twenty-nine percent of fractions delivered were interrupted, of which the Z-axis (superoinferior) motion was the largest contributor to excursion. Baseline corrections significantly compensated for the magnitude of motion in all 3 dimensions (P < .01). The motion relative to the first acquired CBCT was on average seen to consistently increase with treatment time, with the minimum P value occurring at 61.3 minutes. The neural network prediction model was able to predict treatment interruptions with 84% sensitivity on the first 5-minute sample of the trace. Conclusions: Corrections to marker position significantly decreased marker excursions in all 3 axes compared with a single CBCT alignment. Patient-specific modeling may aid in the optimization of cases selected for frameless radiosurgery to increase the accuracy of planned delivery.

Purpose/Objective(s): PBS-PT has a high potential of reducing radiationrelated toxicity in lung and esophagus patients by reducing the dose in the normal tissues surrounding the target. However, potentially highly beneficial steep proton dose gradients can become a disadvantage due to the risk of being misplaced as a result of motion. Therefore, an extensive investigation of the effect of motion on all aspects of proton treatment planning is required. The goal of this study was to use comprehensive 4D imaging data to optimize contouring procedures, to evaluate deformable image registration (DIR) performance, to analyze motion characteristics and to define an optimal treatment planning strategy in preparation for PBS-PT treatments of thorax patients. Materials/Methods: For 20 lung / esophagus patients, weekly repeated 4DCTs were acquired in addition to a 4D planning CT. Furthermore, daily 4DCBCTs were acquired. The influence of different DIR algorithms for contour and dose warping was evaluated. The necessity of manual contour adjustments with respect to differences in accumulated dose and plan robustness was determined. Weekly as well as daily variations in the target motion amplitude were evaluated and the consistency between 4DCT and 4DCBCT information was determined. The frequency of required plan adaptations was compared between 3D optimized proton and VMAT photon plans. Finally, PBS-PT treatment plans optimized in 3D and 4D were evaluated with a comprehensive 4D robustness evaluation method considering the combined influence of setup and range errors, anatomical changes, delivery uncertainties and interplay effects. Results: No difference in the 4D robustness evaluation were found using either of the DIR algorithms. The influence of manual CTV correction on the number of necessary treatment plan adaptations was found minimal (4/ 82 decision changed). Average motion amplitude variations of 2.2mm (range: 0.1-8.8 mm) over the treatment course were found. The median deviation between motions extracted from 4DCT and 4DCBCT was 0.8 mm (range: 0-8.4 mm). 3D optimized proton plans achieve significant heart dose reductions, but require more frequent plan adaptations than VMAT plans. 3D and 4D optimized PBS-PT treatment plans showed to be equally robust against the combination of occurring uncertainties. Conclusion: Comprehensive 4D image datasets were required to gain experience on the impact of motion on different aspects of the treatment planning workflow. Based on the findings, we are defining an efficient and robust treatment planning protocol for PBS-PT treatments of thorax patients. In view of the high incidence of significant motion variations and their potential effects, we are implementing a clinical control infrastructure enabling 4D adapted PBS-PT based on daily motion monitoring.

No abstract

No abstract

Gamma Knife Icon (GKI) enables a user-defined gating threshold for intrafraction motion during stereotactic radiosurgery (SRS). An optimal threshold would ensure dosimetric fidelity of the planned distribution and minimize treatment time extension by gating. A prediction of motion characteristics for a patient based on a retrospective database of motion traces could be beneficial to evaluating the choice of gating threshold. A short acquisition of motion may help to define a personalized threshold that balances dosimetric accuracy and treatment length. This study aims to evaluate the performance of a prediction of motion and the resultant dosimetric consequences for a range of motion gating thresholds. Methods: A database of 2552 motion traces (776 patients) was analyzed using previously published methods to characterize patient intrafraction motion on the GKI. For a selection of six fractionated SRS patient cases (two patients with single brain metastasis, four vestibular schwannomas), a 10-min sample of motion was used to classify motion and identify traces in the database with similar metrics. The similar motion traces were used to perform a predictive reconstruction of the selected patient's delivered dose for a range of motion thresholds. The remaining fractions were reconstructed and compared to that predicted. From the six cases, 26 fractions were used to predict the number of interruptions (n = 26), change in target coverage (n = 26), and change in brainstem maximum dose (vestibular cases only, n = 20). The difference between mean predicted and reconstructed values was compared for accuracy. Results: The difference between mean prediction and reconstructed values was 0.32 ± 0.38% in target coverage, 2.36 ± 5.06 interruptions, and 0.15 ± 0.24 Gy for the brainstem maximum dose. Sixty-seven of the 72 predictions (26 coverage, 26 interruptions, and 20 brainstem maximum dose) were within one standard deviation of the predicted mean. Conclusions: Large databases of motion traces were used to characterize patient performance and predict motion performance. Dosimetric deterioration due to motion and extension of treatment duration can be predicted in some cases using only a short acquisition of motion and the treatment plan. This reconstruction may provide benefit in generating a patient-specific motion threshold.