Objective: Growing evidence supports the role of central nervous system in the modulation of pain in trigeminal neuralgia (TN) patients. The aim of this study was to assess brain functional connectivity alterations in patients with TN before and after neurosurgical treatment of affected trigeminal nerve.

Methods: Sixteen patients with idiopathic/classic TN, who underwent Gamma Knife radiosurgery, were followed up for at least 3 months within an ongoing longitudinal project. They performed clinical and resting-state functional MRI (RS-fMRI) evaluation before and 3 months after treatment. Thirty-three age-and sex-matched healthy controls were also recruited.

Results: Before treatment, TN patients relative to healthy controls showed an increased functional connectivity (FC) (i) of the precentral and postcentral gyrus within the sensorimotor network, (ii) of the right supramarginal gyrus, right postcentral gyrus and bilateral precuneus within the posterior salience network, and an increased FC (iii) of the right fronto-orbital cortex and caudate within the basal ganglia network. Furthermore, a decreased FC of the precuneus, posterior cingulate gyri and lateral occipital cortex within the posterior default mode network was found relative to healthy subjects (Figure 1). Three months after surgery, all patients experienced a significant improvement of facial pain (Barrow Neurological Institute pain intensity score less than IIIb). At postoperative fMRI assessment, no more significant FC increase was found in TN patients relative to controls. On the contrary, a decreased FC of the precuneus and lateral occipital cortex within the posterior default mode network relative to healthy subjects persisted.

Conclusions: In patients with idiopathic or classic TN, pattern of increased FC may reflect the involvement of a system that receives chronic nociceptive stimuli. An effective neurosurgical treatment of the trigeminal nerve appears likewise to modulate and thus reshape abnormal pre-surgical brain circuitries. The study provides novel insights into functional brain alterations of TN patients, which might contribute to disease development and pain change after surgical treatment. In the future, it will be interesting to analyze if specific brain areas can predict the response to neurosurgical treatment.

Objective:  To evaluate the experience of the first center in Brazil, and second in South America, using Contrast Clearance Analysis Software (Brainlab) to differentiate tumor recurrence from radionecrosis in the management of benign and malignant brain lesions after radiation treatments.  

Material and Methods: We analyzed benign and malignant brain lesions (tumors and Arteriovenous malformations) images with Contrast Clearance Analysis (CCA) Software from April 2021 to November 2023 in a Radiation Oncology Center in Brazil. Data from 104 patients and 287 CCA images at our institution were studied. All images were obtained with 3T MRI (Verio, Siemens), and a T1 contrast enhanced volumetric sequence (MPRAGE) was acquired at 5 min and 60 to 105 min. The images were transferred to CCA software. A fusion between the 2 MPRAGE sequences was made and a CCA colored map was calculated. The lesion studied was evaluated according to the color on the map: blue (active tumor) or red (radionecrosis) (Figure 1). The results of CCA software were compared to conventional MRI sequences (diffusion, perfusion and spectroscopy) and in five cases a biopsy was performed.

Results:  Median age was 46.8 years (Range: 4-81) and mean follow up was 29.5 months (Range: 2-57). Patient diagnosis were malignant tumor (89 patients), benign lesions (14) (Figure 5 and 6) and brainstem tumor without biopsy (1). 47,25% patients were treated with single dose radiosurgery, 37,36% with hypofractionation and 15,38% with conventional Radiotherapy. At follow up, 26,3% of patients developed new symptoms and Control MRI with conventional sequences demonstrated disease progression, however, at CCA software was radionecrosis (Figure 2 and 3). 100% had complete symptoms relief after treatment (steroids and vitamin E), and 2 lesions practically disappeared (Figure 4). All biopsied cases were compatible to the CCA software. 

Conclusions:  The CCA software is a new technological approach providing efficient distinction between tumor/ radionecrosis. The methodology provides high resolution and easy to interpret images with high accuracy. The present study is the first to describe the CCA software contribution among benign tumors and AVMs. 

Purpose: Gamma Knife stereotactic radiosurgery (GKRS) is a widely used treatment for pituitary adenomas (PAs) due to its high precision and efficacy. However, the volume response of PAs to GKRS varies among patients, underscoring the importance of identifying reliable predictors of treatment outcomes. Radiomics, a quantitative imaging analysis approach, can potentially extract imaging biomarkers that can aid in predicting treatment response. This study aims to pioneer the use of radiomic MRI analysis for predicting pituitary adenoma volumetric response to GKRS.

Methods: A comprehensive radiomics analysis was performed to predict the volume response of PA to GKRS. The retrospective cohort consisted of 80 patients who underwent GKRS for 29 functional PA and 51 non-functional PA. Forty-eight patients were treated with a single dose of 12 to 40 Gy to the PA margin, and 32 patients were treated with a hypofractionated regimen of 3-5 fractions with 11-35 Gy single fraction equivalent dose (SFED). After a follow-up period of 40.4 (7 - 106) months, a total of 98.8 % of PAs were controlled. The volumetric tumor change varied widely between 90% regression and 93% progression, with a mean regression of 45.7%. Pre-treatment T1w, T2w, FLAIR, and CE-T1w sequences acquired with 3-Tesla MRI were used to extract 2156 radiomic features that captured the tumors' intensity, shape, and texture characteristics. Radiomic signatures were generated using the least absolute shrinkage and selection operator (LASSO) for feature selection, in conjunction with several classifiers: random forest, naïve Bayes, kNN, logistic regression, neural network, and SVM.

Results: The models demonstrated predictive performance in the validation folds with AUC values ranging from 0.759 to 0.928 and R2 values between 0.272 and 0.665. Single-sequence T1w, dual-sequence T1w+CE-T1w, and multi-modality including clinicopathological (CP) characteristics (CP+T1w+CE-T1w) achieved similar prognostic performance in validation folds, with respective AUCs of 0.928, 0.899, and 0.909. All these radiomics models significantly (t-test) outperformed a benchmark model involving only clinicopathological features (AUC=0.846). The single-sequence model, including only CE-T1w features, provided the weakest prognostic performance with an AUC of 0.759.

Conclusion: This study is the initial radiomic analysis aimed at predicting the volume response of PAs to GKRS. Notably, the developed MRI-based radiomics models exhibited better classification performance compared to the benchmark model composed only of standard clinicopathological parameters. The clinical significance of this result is based on its potential to enable the individualization of therapeutic strategies, thereby enhancing treatment outcomes.

Purpose

Automatic contouring of brain metastatic target volumes may improve the efficiency of stereotactic radiosurgery (SRS) workflows. Generating clinically acceptable contours is necessary to realizing this potential. This work aims to evaluate the accuracy of an AI contouring system for brain metastasis (BM) gross tumors volumes (GTVs) with respect to contours defined by physicians.

Methods

Post-contrast T1-weighted MR images and GTVs of BM were retrospectively collected from 2092 patients treated using SRS at seven institutions. Centralized data curation was done by two radiologists to delineate GTVs of uncontoured metastases (present in the brains but not amongst the treated tumors). An automated ‘artificial intelligence’ (AI) system based on nnU-Net with adaptive Dice loss and synthetic data augmentation was trained (N=1907) and evaluated (N=185) on non-overlapping subsets of the data. A second testing subset (N=206) was evaluated after the completion of training. BM detection was assessed by sensitivity and false positive rate (FPR). The operation points of the AI system were selected to achieve the target sensitivity of 0.9. To assess interobserver contouring variability, three clinicians each contoured 163 selected BM GTVs from 20 testing patients; one lesion was excluded due to user error during annotation. Interobserver contouring variability between clinicians was quantified as the average pairwise values of three contouring metrics (Dice similarity coefficient (DSC), 95-percentile Hausdorff distance (HD95) and average HD (AHD)). The AI system contouring agreement was compared by measuring the same metrics between the system and each clinician.

Results

The AI system achieved overall BM-level detection sensitivity of 0.904 at an FPR of 0.65±1.17 on the first testing dataset, and sensitivity of 0.907 at an FP rate of 0.57±0.8 on the second dataset. Mean values of DSC, HD95 and AHD were 0.758, 1.45 mm and 0.23 mm, respectively, for the first test set and 0.705, 1.91 mm and 0.33 mm, for the second. On the interobserver variability of 20 patients, clinician-clinician mean DSC, HD95 and AHD were 0.714, 1.32 mm and 0.25 mm, respectively. The AI-clinician mean DSC, HD95 and AHD were 0.739, 1.23 mm and 0.22 mm, respectively.

Conclusion

The AI system showed contouring variability from clinician contours on par with interobserver variability. Further evaluation will be carried out to evaluate the AI system’s clinical utility.

Introduction: Prostate-specific membrane antigen targeted molecular imaging with positron emission tomography (PSMA/PET) is being increasingly utilized in care of prostate cancer patients, allowing for metastatic directed therapies as well as PSMA targeted radionucleotide therapies with significant survival benefits. We present a case series of patients with intra and extra-axial brain metastasis detected on PSMA/PET imaging treated with brain stereotactic radiosurgery (SRS).

Methods: A prospectively collected database was queried for prostate cancer patients who underwent brain SRS from 1/2020 to 12/2023 in an NCI-designated Comprehensive Cancer Center. The patients who underwent F18-PSMA/PET imaging preceding SRS were identified and their clinical course and imaging findings were reviewed.

Results: Among fourteen prostate cancer patients who had received brain SRS, five had undergone PSMA/PET imaging in biochemical recurrent setting, yielding new diagnosis of brain metastases. Two patients had received Lutetium-177 PSMA therapy (Pluvicto) for metastatic skeletal disease a priori. At the time of PSMA/PET imaging, median age was 64years (range: 49-70). Two patients were asymptomatic, two had headaches and one suffered from blurred vision. Two patients had single brain metastasis and three had multiple, with mean standardized uptake values (SUV) of 13 (range 11-19.5) on PSMA/PET. Three also had calvarium/dural based lesions. Brain MRI was acquired for all patients and showed corresponding findings, although extent of base of skull involvement was better delineated with PSMA/PET. Median time to SRS from PSMA/PET imaging acquisition was 3weeks (range 0-15) and one patient had post-SRS resection. Median dose of 16Gy (range 12-21) in 1 fraction was delivered utilizing frame-less gammaknife radiosurgery. Patients were followed with subsequent PSMA/PET and/or brain MRIs. No in-field recurrences were seen in median follow up of 5months (range 2-15), three required SRS to other brain lesions and one died.

Discussion: Brain metastases from prostate cancer are rare though can be effectively treated with SRS. PSMA/PET imaging combined with brain MRI allows for increased sensitivity and specificity of brain metastasis detection and radiosurgery target delineation. Combination of radionucleotide therapy and brain SRS appears safe.

Purpose/Objectives:

            Magnetic resonance (MR) imaging is increasingly integrated into Radiation Oncology (RO) departments with the development of MR-linacs and MR simulation. MR technology allows for better soft tissue contrast which is important for precise tumor delineation in stereotactic radiosurgery (SRS) and stereotactic body radiotherapy (SBRT). Due to the number of foreign bodies and metal implants in patients with cancer, adoption of a comprehensive patient screening and MR safety workflow in RO is critical. Identifying MR unsafe implants only at the time of MR simulation leads to same-day cancellations, potentially delaying treatment, and can risk MR safety events (SEs).

Materials/Methods: 

In an effort to decrease same-day cancellations and improve safety of a 3-Tesla MR simulator, three plan-do-study-act (PDSA) cycles were implemented from 4/18/22 – 1/19/23. PDSA cycle 1 involved implementation of a two-screen functional workflow, adapted from radiology at the same institution. PDSA cycle 2 and 3 involved education for stakeholders. PDSA cycle 3 educational intervention included a visual aide to assist with work queue (WQ) use. Endpoints evaluated included the number of same-day cancellations, patients in the WQ (a measure of the number of patients identified at the initial screen as having an implant), and SEs in each PDSA cycle.

Results: 

PDSA cycle 1 spanned 56 workdays during which 91 MR simulations were scheduled with 6 cancellations (6.5%). PDSA cycle 2 spanned 84 days during which 173 MR simulations were scheduled with 18 cancellations (10.4%). PDSA cycle 3 spanned 39 workdays and had 94 MR simulations, with 7 cancellations (7.4%). The cancellation rate during each PDSA cycle was 0.11, 0.21, and 0.17 cancellations/day, respectively. The number of patients in the WQ during each PDSA cycle, representing successfully screened high-risk patients, was 0, 0, and 3, respectively. There were no SEs during the study.

Conclusion: 

In this study, an MR safety workflow from radiology was successfully implemented into a RO department. There were no SEs during the study, but the number of patients successfully screened as high-risk and placed in the WQ increased after repeat MR education. Further increases in WQ use would decrease the demand for implant assessment at point of care, which could decrease burden on the MR technologist, reduce same day cancellations, and potentially SEs. As the demand for MR simulation for stereotactic radiation target delineation increases, repeated continued and updated MR specific education is important to increase efficiency of MR simulation appointments and maintain patient safety.

Purpose/Objective:
Stereotactic radiotherapy (SRT) is a promising treatment option for patients with multiple brain metastases. Optimal treatment quality with sparing of healthy brain tissue is essential to avoid SRT complications such as brain necrosis. The aim of this work was to evaluate the quality and the dosimetric accuracy of one-isocenter treatment plans using film measurements.

Material and methods:

28 patients with multiple brain metastases median= 7 (4-18) with a total of 244 lesions were treated with SRT (8 - 25.5 Gy in 1 or 3 fractions). Non-coplanar LINAC-based treatment planning was performed in RayStation with 6 MV FFF. One-isocenter treatment technique consisted of 6 VMAT arcs with 3 different couch rotations (0°, 60° and 300°). For each arc, automatic optimisation of a collimator angle was used. The used gross tumour volume to planning target volume (PTV) margin was 1 mm. The total PTV ranged from 1.06 till 60.8 cm³ (median= 13.1 cm³). The PTV coverage was at least 99%. To assess SRT plan quality, the Paddick conformity index (CI), the Paddick gradient index (GI), the total V_12Gy of uninvolved brain, the number of monitor units and irradiation time were studied and reported as MEAN±1SD.

Treatment plans were recalculated on a home-made phantom consisting of 4 layers of plastic. In between the layers, GafChromic EBT3 films were placed and after 24 hour analysed using DoseLab 4.11. Gamma pass rate was calculated with 3% absolute global dose difference and 1 mm distance-to-agreement criteria. Additionally, dose calculations in Mobius3D were performed with the same gamma criteria.

Results:

The results averaged over 28 patients are reported. The mean total CI and GI were 0.73±0.10 and 5.35±2.05, respectively. The total V_12Gy was 5.31±5.56%. The number of monitor units and the irradiation time were 6057±2505 MU and 597±56 seconds, respectively. Additionally, we studied the CI and GI as a function of the total volume (see figure 1).

The mean pass rate for all measured films was 94.6±0.4% with 3% and 1 mm distance-to-agreement criteria. The mean pass rate of the dose calculations in Mobius was 98.2±1.0% with the same criteria.

Conclusion:

The dosimetric results obtained with the film measurements and secondary dose calculations are in good agreement within our criteria. High plan quality was observed with a dependence on the total volume of the metastases: increasing the CI and decreasing the GI. These results give confidence in the accuracy of the method and our treatment plans.

Introduction

Frameless stereotactic radiosurgery (SRS), facilitated by image-guided positioning systems, is a widely accepted treatment modality for both benign and malignant brain lesions. In the context of arteriovenous malformations (AVM), digitally subtracted angiography (DSA) offers high temporal resolution and dynamic flow information. However, the process of defining the target involves the registration of orthogonal two-dimensional (2D) DSA views of the AVM nidus alongside multimodality three-dimensional (3D) imaging. This necessity causes a delay in the seamless integration of the frameless approach, primarily due to the need for a dedicated image localizer to immobilize the head under rigid and invasive frame fixation to the skull. This study was aimed to evaluate the targeting accuracy achieved by a dedicated software, Elements Angio (Brainlab AG, Munchen, Germany), which enables a frameless procedure. 

Material and Methods

A retrospective comparison was conducted on ten patients previously treated using our established frame-based SRS at our institution. Datasets from DSA in both coronal and sagittal planes, as well as magnetic resonance angiography (MRA), were imported into the Elements software for targeting the nidus. From the MRA image series, a 3D vasculature tree containing vascular details was automatically extracted. Subsequently, it was manually and automatically co-registered to a selected frame pair of 2D DSA vascular images using a six-degree-of-freedom rigid registration. Target delineation for SRS planning followed, and the resulting structure was compared to the clinically reference contoured structure used in frame-based SRS treatment. The evaluation of similarity between target contours employed several metrics, including the Dice Similarity Coefficient (DSC), Jaccard Index (JI), Hausdorff distance (HD), and Mean Distance to Agreement (MDA).

Results

No significant difference in AVM nidus volume was found between frame-based and frameless approach (0.79 ± 1.35 cc and 0.83 ± 0.99 cc, respectively). Comparable targeting objects by means of DSC and JI were found: 0.74 ± 0.08 and 0.59 ± 0.10, respectively. Sub-millimetric MDA was found (0.67 ± 0.32 mm). The mean HDshift between individually contoured volumes was 2.97 ± 1.33 mm. These differences were related to the small differences in contouring and segmentation of the objects. 

Conclusion

The study supports the viability of the frameless approach based on vasculature registration using Elements Angio, emphasizing its precise accuracy in target localization and its clinical applicability, offering the advantage of eliminating the need for invasive angiography on the day of SRS.

Purpose: Workflow changes and verification techniques in mask fixation over the last five years at our institution have led to an increase in the percentage of mask- based procedures.  The Leksell Gamma Knife ICON system (LGKS) CBCT has the ability to correct for rotational and translational inter-fractional motion. We verified the rotational and translational repositioning accuracy of the LGKS with a prototype 6 degree of motion platform, an SRS MapCheck®, StereoPHAN™, and 3D printed accessories and an anthropomorphic phantom with gafchromic film.

Methods: From the XML log files continuous tracking of each patient can be reconstructed to determine the net x,y,z motion of the infrared sticker placed on the patient's nose used by the HDMM to determine intrafractional motion.  Subsequent repositioning CBCTs after the sticker has moved out of tolerance were used to determine the relative motion of the tumor or delivered shots relative to the sticker.  Repositioning accuracy was tested by acquiring an initial CBCT defining the stereotactic reference of the phantom positioned a known distance from the origin of the LGKS coordinate system. Rotational shifts of 1.00, 5.00, and 10.00 +/- 0.01 degrees were applied to the phantom before acquiring a subsequent CBCT. This was registered to the stereotactic reference for angular repositioning accuracy. To test for repositioning accuracy of shot delivery, a plan was spatially separated shots and delivered with and without rotation to the SRS. The rotational corrections were analyzed for consistency of delivered dose. Additionally, translational shifts were applied verifying treatment interruption beyond set 3mm limit and the accuracy of the High Definition Motion Management system and translational repositioning accuracy.

Results: Rotational repositioning accuracy was tested for roll, pitch and yaw with maximum registration differences of 0.09, 0.03, 0.08 degrees. Translational repositioning was verified for shifts up to 1cm with an accuracy of 0.2mm. Registration was performed with matching window levels and a tight region of interest (ROI) on anterior edge of the cube. Treatment delivery of rotationally corrected shots were within 3% of standard. 

Conclusion:  Conclusion: With some example calculations of tumor motion relative to the patients nose and verification of rotational repositioning accuracy of the LGKS CBCT system we were able to confidently increase the frameless workload at NYU from 3% to 23%, mainly for metastases cases.  Mask fixation commonly decreased the amount of overall time required by the patient, and staff, although some patients have movement challenges that extend care times.

Background

Single-isocenter linac-based stereotactic radiosurgery (SRS) is established as a treatment modality for multiple brain metastases. This technique may exhibit an increased sensitivity to rotational and translational errors. Consequently, image-guidance for motion management has become crucial, providing an improved certainty in patient positioning. The purpose of this study was to analyze intra-fraction positioning errors measured during beam delivery, their impact on the dose administered and to re-evaluate the margins used to compensate for positioning uncertainty.

Materials and methods

33 consecutive patients corresponding to 127 brain lesions and 356 treatment arcs were included retrospectively. Treatments were planned with non- coplanar dynamic conformal arcs using both a margin and a no-margin strategy on the targets. Intra-arc positioning errors were measured using stereoscopic x-rays (ExacTrac Dynamic, Brainlab), triggered by the gantry position during arc delivery. Couch corrections above 0.7mm and 0.5° were always applied. Intra-arc positioning data was analyzed. The impact of positioning errors on the dose was evaluated by applying the measured errors on each arc to their correspondent dose distributions and reconstructing a realistic delivered dose.

Results

Median residual errors were 0.10mm, 0.13mm and 0.08mm for the lateral, longitudinal and vertical directions and 0.10°, 0.08° and 0.13° for the pitch, roll and yaw angles respectively. 90% of the treatment arcs showed shifts of less than 0.4mm and 0.4°in all directions. 3D displacement was found lower than 1mm for 99% of the arcs studied. Dosimetric impact of motion showed the largest losses in coverage on small targets. Prescription coverage to at least 98% of the target was fulfilled by 92.1% of the targets planned with 2mm margin and by 81.9% of the targets planned with no margin. All targets achieved at least 95% of the prescription dose to 95% of their volume, even when planned without margins. No correlation was found between coverage loss and lesion-to-isocenter distance.

Conclusions

Intra-fractional errors measured during beam delivery were found to be notably low with a dose impact that showed acceptable target coverage when applying these intra-arc errors to the dose distributions of the individual treatment arcs. Using an adequate immobilization and intra-fraction monitoring imaging prior to and during irradiation, and applying a strict beam-hold strategy to prevent positioning errors, no margins need to be added to compensate for intra-fraction motion. For very small lesions (<0.1cc), the addition of a margin may still be of benefit.

Objective:

In Linac-based SRS, small field condition and modulation complexity pose significant challenges to the modeling accuracy of multi-leaf collimator(MLC). In current Eclipse planning system, a binary MLC model with tunable dosimetric leaf gap(DLG) is used. For high precision SRS treatments, unsatisfactory agreement between measured and calculated doses is reported, and a trial-and-error tuning of DLG is often needed, leading to increased commissioning complexity and user-dependent variability. In the latest version of Eclipse(v18), an enhanced MLC model is implemented by constructing the actual rounded-end leaf design. We intended to investigate the efficiency and accuracy of the enhanced leaf model(ELM).

Methods: 

Dose calculations using Analytical Anisotropic Algorithm(AAA) were performed in a test Eclipse v18 environment, with original beam data from an Edge linear accelerator with HD-MLC. The DLG for 10FFF was optimized for SRS treatments, denoted as AAA-SRS-16. For 6FFF, there was no clinical SRS beam model, so the original AAA-16 model was used. The ELM parameters were measured by solid water and PTW N30004 ion chamber, and configured in the test system, named AAA-18. The doses were re-calculated using AAA-18 for(1) six static on-axis small fields from 0.5x0.5cm2 to 4x4cm2 and one off-axis small field of 1x1cm2 and up to 6cm from isocenter, (2)six single-isocenter single target HyperArc plans. Gafchromic EBT4 film was used for measurement. 

Results: 

The time spent on ELM parameters measurement was comparable to the original MLC measurement. However, substantial time saving was associated with the ELM configuration. In the original SRS modeling, we re-tuned DLG four times to find the value with acceptable agreement with measurement, while each round took hours for re-calculation and re-evaluation. The ELM configuration was one-time implementation and only required one round of calculation and verification on the same dataset. 

AAA-18 demonstrated comparable or improved agreement with measurements. For static small fields, the average gamma-passing rate at 3%/1mm 10% threshold is 99.8% for AAA-18 and 98% for AAA -16 for 6FFF, and 97.5% for AAA-18 and 98% for AAA-SRS-16 for 10FFF. For patient plans, both models achieved 100% 3%/1mm passing-rate and 99.5% 1%/1mm passing-rate at 10FFF. At 6FFF, we observed an improvement of 3%/1mm and 1%/1mm of AAA-18, at 100% and 98.3%, compared with 98.7% and 92.1% of AAA-16.

Conclusion: 

The new enhanced leaf model introduced in Eclipse v18 substantially improves efficiency and consistency of modeling process of the Eclipse dose calculation algorithm while maintaining comparable or superior accuracy for Linac-based SRS.

Objective: The Leksell Gamma Knife (LGK) Icon™ offers frameless (mask-based) fixation with its high-definition motion management (HDMM) system. However, HDMM only records intra-fraction motion values of the nose marker without detailing changes in the X, Y, and Z axes. This study investigates the coordinate shifts of nose marker using the HDMM system during frameless gamma knife radiosurgery (GKRS) for meningioma and metastases patients.

Methods: We retrospectively studied patients diagnosed with meningioma or metastases who underwent frameless GKRS using the LGK Icon^TM from January to September 2023. All patients were immobilized using a Nanor thermoplastic mask (Orfit Industries, Wijnegem, Belgium) for frameless fixation. We imported sequential data of the nose marker coordinates (X, Y, and Z) into the LGK system, and then analyzed the subsequent coordinate changes

Results: We evaluated patients with meningioma (n=18) and metastases (n=30) treated with GKRS under frameless fixation. All patients exhibited no cognitive impairments and remained compliant. The median beam-on time was 64.5 minutes for meningioma patients and 93.75 minutes for metastases patients. In meningioma patients, there were no significant differences in the X (0.07 ± 0.06 mm), Y (0.08 ± 0.46 mm), and Z (0.08 ± 0.45 mm) axes. However, metastases patients showed significant differences between the X (0.33 ± 0.23 mm) and Y (0.57 ± 0.37 mm, *p<0.05) axes. No significant differences were observed between the X and Z (0.43 ± 0.31 mm) axes or between the Y and Z axes. An analysis of the movement over time revealed a significant increase in the Y-axis movement after 30 minutes.

Conclusion: The Y-axis movement, as indicated by the HDMM, was most prominent in patients with metastases. We recommend pressing the upper part of the nose when securing the mask to minimize nose marker movement in the Y-axis. Furthermore, when creating treatment plans for patients with metastases using the LGK Icon™, we suggest adding a margin of approximately 0.5 mm in the Y-axis.

Objective: In recent years, stereotactic radiosurgery (SRS) has gained an increasing role in controlling recurrence or progression of atypical World Health Organization (WHO) grade 2 meningiomas. This study evaluates a large, multi-institutional database of European Radiosurgery centers to present the outcomes of WHO grade 2 meningiomas treated with SRS. Our aim was to investigate the long-term tumor control rate and durable morbidity of SRS.

Methods: At 16 participating centers, 355 consecutive patients with 593 WHO grade 2 meningiomas treated between 1992 and 2022 were enrolled in the study. Clinical and imaging data were collected by each center and uniformly entered into a multicenter database.

Results: Detailed results of 559 meningiomas (94%) were analyzed. The median age of the patients upon SRS treatment was 56 years. The median tumor volume was 5,7 cm³ and the median dose at the tumor margin was 16 Gy (50% isodose). All tumors were treated by surgery before SRS. The median follow-up time at imaging was 46 months. The three and five-year progression-free survival (PFS) rate was 66.0% and 47% respectively. The three and five-year local tumor control (LC) rate was 74.0% and 54% respectively.  The morbidity rate at last follow-up was 9.9%. 

Conclusion: We show that SRS for previously resected WHO grade 2 meningiomas has a reasonable tumor control rate that also compares favorably with the literature in the medium to long term. We also observed a low morbidity rate. Our study suggests that SRS may be an effective option for controlling WHO grade 2 meningiomas. However, it should be noted that meningiomas can vary widely in their behavior and response to treatment, and the best approach depends on the specific characteristics of the tumor and the individual patient's situation.

Objective: Cranial Nerve Neuropathies (CNNs) frequently manifest in Cavernous Sinus Meningioma (CSM). Stereotactic radiosurgery (SRS) or fractionated stereotactic radiosurgery (FSR) are well-established upfront treatments for CSM. This study aims to assess the rates of recovery, time-to-improvement, and recovery patterns of CNNs in patients with CSM who have undergone  SRS or FSR. The reported outcomes were analyzed to gain insight into the efficacy of these treatment modalities.

Methods: A retrospective study was conducted on patients with CSM treated with LINAC-based SRS/FSR between the years 2005-2020 at a single institution. A total of 128 treated patients were treated during this period, with 46 patients presenting with CNNs. The study collected and analyzed patient demographics, clinical parameters, SRS/FSR treatment characteristics, post-treatment CNNs recovery duration, status, and radiological control on the last follow-up magnetic resonance imaging (MRI) scan. 

Results: The median follow-up duration was 53.4 (range, 3.9-190.4) months. The mean age at diagnosis was 51.8 (range, 19.1-75.7) years. SRS was performed on 25 patients and FSR was performed on 21 patients. The mean pretreatment tumor volume was 9.5 cc while the mean end-of-follow-up tumor volume was 5.1 cc. The mean marginal dose was 12.8 and 48.7 Gy for SRS and FSR, respectively. CNNs recovery was documented in 80.4% of the patients with extra-ocular CNNs showing improvement in 43.2% of the patients, trigeminal CNNs in 32.4%, and optic CNNs in 10.8%. The median time-to-improvement was 3.67 months, with FSR showing a longer time-to-improvement (12.9 months) compared to SRS (2.5 months, p=0.04). The radiological tumor control rate was 100%. 

Conclusions: This study suggests that SRS/FSR for CSM provides good and sustainable CNNs recovery outcomes with excellent long-term radiological control. A higher CNNs recovery rate was associated with a smaller pre-treatment tumor volume, while shorter time-to-improvement was identified in patients treated with SRS compared to FSR, particularly in those with small pre-treatment tumor volume.

Background: This research was performed to determine the utility of the addition of [68Ga]-DOTATATE PET imaging to MRI in meningioma response assessment following radiosurgery.

Methods: Radiosurgery (SRS) for 27 patients with 64 meningiomas was planned using co-registered DOTATATE PET/MRI. 7 patients (26%) had WHO grade 1 meningiomas, 11 patients (41%) had WHO grade 2 meningiomas, and 7 patients (26%) had WHO grade 3 meningiomas. A single patient (4%) did not have a diagnostic biopsy that permitted WHO grade determination. 26 patients (96%) were treated with fractionated SRS, and one (4%) received single-fraction SRS. The mean and modal SRS doses were 30 and 35 Gy, respectively, and the mean and modal fraction number was 5. For all patients, follow-up DOTATATE PET/MRI was performed at 6-12 months post SRS. The maximum absolute standardized uptake value (SUV) and SUV ratio (SUVR_SSS) referencing the superior sagittal sinus blood pool were measured for the DOTATATE scans. Size change on MRI was determined by RANO criteria. The association of SUVR_SSS change magnitude and PFS was evaluated using Cox regression.

Results: Each patient served as her/his own control. Post-irradiation SUV and SUVR_SSS decreased by 37.4% and 44.4%, respectively (p < 0.0001). Size product decreased by 8.9%, thus failing to reach the 25% significance threshold as determined by RANO guidelines. Mean follow-up time was 26 months (range: 6-44). Mean PFS at 34 months was 83% overall, and 100%/100%/54% in WHO-1/-2/-3 subcohorts, respectively. At last follow-up (42-44 months), PFS was 83%/100%/54% in WHO-1/-2/-3 subcohorts, respectively. Cox regression analyses revealed a hazard ratio of 0.48 for a 10-unit reduction in SUVR_SSS in patients treated with SRS.

Conclusions: DOTATATE PET SUV and SUVR_SSS demonstrated marked and significant decreases following SRS. Lesion size decrease on MRI after SRS was statistically significant, however it did not reach clinical significance by RANO criteria. DOTATATE PET/MR thus represents a promising imaging biomarker for response assessment in meningiomas treated with radiosurgery.

BACKGROUND: New-onset of vestibular symptoms (Vsym), such as dizziness and imbalance, can be side effects after stereotactic radiosurgery (SRS) for vestibular schwannomas (VS). Although these Vsym can severely affect the daily life of VS patients, there are limited data available providing prognostic information on the risk of developing VD after SRS and dose constraints for SRS of VS.
METHODS: We included patients who received Cyberknife® SRS for newly diagnosed unilateral VS between 2012 and 2022 and a minimum of two follow-up (FU) visits. The incidence of vestibular symptoms before and after treatment was recorded and correlated with tumor-, patient-, and treatment related characteristics.
RESULTS: We identified 205 patients with a median age of 58 years (range: 20-83) and a median follow-up of 37 months (range: 12-105). Mean tumor volume was 1.4 cm³ ± 1.5 ml (range: 0.07-8.6). A mean marginal dose of 13 Gy ± 0.2 (range: 12-14) was administered to the tumors. The mean Dmax to the vestibule was 8.5 ± 2.9 (range: 2-14.6). Treatment failure was noted in two patients (0.9%). About 68 % (n=141) of the patients had Vsym prior to SRS. In 38% (n=54) of these patients Vsym improved after SRS. Of the remaining 64 patients who did not have Vsym before treatment, 24 (37%) developed new Vsym (vertigo, n=4; balance disorders, n=6; mixture, n=14). The median time to onset of symptoms was 6 months (range: 4-37). In half of patients the new symptoms completely resolved within a median time of 7.5 months (range: 5-36). Kaplan-Meier estimates the probability of new onset of permanent Vsym with 15% after one and 17% after two years. In multivariate analysis, neither tumor volume (p=0.6), age (p=0.2), nor radiation dose to the vestibule were significantly associated with the occurrence of Vsym.
CONCLUSIONS: SRS leads to an improvement of Vsym in more than one third of this cohort. Moreover, SRS resulted in preservation of vestibular function with more than 80%. The incidence of Vsym after SRS was found to be independent from the dose exposure to the vestibular apparatus.

Purpose/Objectives HyperArc™ (HA) by Varian Medical Systems is a treatment planning system developed in 2017 that automates both planning and delivery of single-isocenter VMAT radiosurgery (SRS). HA was intended for complex multi-metastasis cases, for which it generates high-quality, rapidly-deliverable plans. The effectiveness of treating benign intracranial tumors (BIT) with HA was unknown. With much longer life expectancy than that of multiple malignant brain metastases and slower tumor growth, it is arguably more imperative to create high-quality, safe radiosurgery plans for BIT. We have prospectively collected data on treatment planning, delivery, and clinical outcomes of BIT managed with SRS since HA deployment. We report a two-year update on all outcomes.

Materials/Methods Patients included received SRS between 2018 and 2021 using HA. WHO grade 3 meningiomas were excluded. Full prescription dose was normalized to ≥ 99% of gross tumor volume without additional expansion. Treatments were delivered on Varian Edge linear accelerator with 10MV flattening-filter free beam at 2400 MU/min with high-definition multi-leaf collimator. Post-treatment imaging, toxicities, and standard pathology-specific outcomes were assessed at follow-up. Significant CNS toxicity defined as grade 3 or higher by Common Terminology Criteria for Adverse Events (v5.0).

Results 198 BIT targets in 183 patients were treated with 186 HA plans. Most common pathologies were meningiomas (122), pituitary adenomas (30), and acoustic schwannomas (23). Nearly half (45.2%) were treated in a single fraction (12-22Gy), 54.8% were treated with fractionated SRS (24-35Gy). Mean RTOG CI and Paddick GI were 1.12 and 3.31, respectively.  A majority (75%) were treated with 3 arcs with mean total treatment and beam-on time lasting 10.5 and 2.2 minutes, respectively. Mean FU was 2.6 years. 20 of 198 (10.1%) tumors progressed with mean time to failure being 2.1 years. Of those that progressed, 19 were meningiomas of which 17 were WHO grade 2 with prior surgery. Significant CNS toxicity was reported in 7.0% of patients, most of whom had cerebral edema requiring medical/surgical intervention. Of those with BIT near cochlea, 3.4% did not preserve hearing. Of those near the optics, long term visual preservation was 98.6%.

Conclusions This is the largest registry with longest follow-up of benign intracranial tumors treated with HA to date. HA continues to efficiently deliver high quality plans with sharp dose fall-off for BIT. Benefits of treatment being completed within minutes may include improved patient experience, decreased intrafractional motion, and more efficient use of resources. Clinical outcomes remain congruent with historical controls. 

Methods: One hundred out of 142 consecutive eligible patients with a newly diagnosed VS with CPA diameter

Results:  Two patients withdrew after randomization, the remaining 98 completed the 4-year trial and were included in the primary analysis (mean age 54 years, 42% female). For the primary outcome, the study showed a significant difference; V4/V0 was 0.87 in the SRS group, and 1, 51 in the expectation group (p<0.002). In the upfront SRS group, 3 patients received additional treatment; one had additional SRS, two underwent surgery. The remaining 45 patients in this group did not require any additional treatment. In the expectation group, 23 patients received treatment due to tumor growth, 21 with SRS, 1 with surgery. The remaining 27 did not receive any treatment during the study period. The comparative analysis of the remaining 25 secondary outcomes including hearing, did not reveal any between-group differences.

Conclusion: The study provides class 1 evidence for the growth inhibiting effect of GKRS on vestibular schwannoma, and suggests that the treatment does not lead to any other beneficial effects over a 4-year period. Notably, 27 out of 50 patients in the observational group did not require any treatment during the study period due to the quiescent nature of their tumor.

Reference: Dhayalan D et al, JAMA, August 1 2023; 330(5), 421-31.