Courses Of Stereotactic Radiosurgery Research Articles

Border Zone Maybe Correlated with Radiation Necrosis After Radiosurgery in Metastatic Brain Tumor

BackgroundRadiation necrosis (RN) after stereotactic radiosurgery (SRS) in brain metastases has been extensively evaluated, and RN is correlated with various risk factors. However, no study comprehensively analyzed the correlation between RN and the border zones of the brain that are vulnerable to ischemia. We hypothesized that patients with tumors in the border zone are at high risk of RN. Hence, the current study aimed to assess the correlation between border zone lesions and RN, with consideration of other predetermined factors. MethodsThis retrospective study included 117 patients with 290 lesions who underwent Gamma Knife® stereotactic radiosurgery (SRS). Radiological and clinical analyses were performed to identify factors possibly correlated with RN. Notably, the lesion location was classified into two groups (border zone and non-border zone) based on the blood supply. ResultsIn total, 22 (18.8%) patients with 22 (7.5%) lesions developed RN. Univariate analysis revealed a significant correlation between RN and external border zone lesions, second course of SRS administered at the same site of the previous SRS, prescribed dose, and tumor volume. Multivariate analysis showed that border zone lesions, second course of SRS at the same site of the previous SRS, and tumor volume were significantly correlated with RN. ConclusionsPatients with tumors in the border zone are at high risk of RN. The potential risks of RN can be attributed hypothetically to hypoperfusion. Hence, the association between RN and border zone lesions seems reasonable.

Determinants of Symptomatic Intracranial Progression After an Initial Stereotactic Radiosurgery Course

Purpose: Clinical and imaging surveillance of patients with brain metastases (BMs) is important after stereotactic radiosurgery (SRS), as many will experience intracranial progression (ITCP) requiring multidisciplinary management. The prognostic significance of neurologic symptoms at the time of ITCP is poorly understood. Materials/MethodsThis was a multi-institutional, retrospective cohort study from 2015-2020, including all patients with BMs completing an initial course of SRS. The primary outcome was overall survival (OS) by presence of neurologic symptoms at ITCP. OS, freedom from ITCP (FF-ITCP), and freedom from symptomatic ITCP (FF-SITCP) were assessed via Kaplan-Meier method. Cox proportional hazard models tested parameters impacting FF-ITCP and FF-SITCP. Results: Among 1383 patients, median age was 63.4 years, 55% were female, and common primaries were non-small cell lung (49%), breast (15%), and melanoma (9%). At a median follow up of 8.72 months, asymptomatic and symptomatic ITCP were observed in 504 (36%) and 194 (14%) patients respectively. The majority of ITCP were distant ITCP (79.5%). OS was worse with SITCP (median 10.2 vs 17.9 months, p<0.001). SITCP was associated with clinical factors including total treatment volume (P = 0.012), melanoma histology (P = 0.001), prior WBRT (P = 0.003), number of brain metastases (P < 0.001), interval of 1-2 years from primary and brain metastasis diagnosis (P = 0.012), controlled extracranial disease (P = 0.042), and receipt of pre-SRS chemotherapy (P = 0.015). Patients who were younger and received post-SRS chemotherapy (P = 0.001), immunotherapy (P < 0.001), and targeted or small-molecule inhibitor therapy (P < 0.026) had better FF-SITCP. Conclusions: In this cohort study of BM patients completing SRS, neurologic symptoms at ITCP is prognostic for OS. This data informs post-SRS surveillance in clinical practice as well as future prospective studies needed in the modern management of BMs.

Incidence and imaging characteristics of difficult to detect retrospectively identified brain metastases in patients receiving repeat courses of stereotactic radiosurgery.

During stereotactic radiosurgery (SRS) planning for brain metastases (BM), brain MRIs are reviewed to select appropriate targets based on radiographic characteristics. Some BM are difficult to detect and/or definitively identify and may go untreated initially, only to become apparent on future imaging. We hypothesized that in patients receiving multiple courses of SRS, reviewing the initial planning MRI would reveal early evidence of lesions that developed into metastases requiring SRS. Patients undergoing two or more courses of SRS to BM within 6months between 2016 and 2018 were included in this single-institution, retrospective study. Brain MRIs from the initial course were reviewed for lesions at the same location as subsequently treated metastases; if present, this lesion was classified as a "retrospectively identified metastasis" or RIM. RIMs were subcategorized as meeting or not meeting diagnostic imaging criteria for BM (+ DC or -DC, respectively). Among 683 patients undergoing 923 SRS courses, 98 patients met inclusion criteria. There were 115 repeat courses of SRS, with 345 treated metastases in the subsequent course, 128 of which were associated with RIMs found in a prior MRI. 58% of RIMs were + DC. 17 (15%) of subsequent courses consisted solely of metastases associated with + DC RIMs. Radiographic evidence of brain metastases requiring future treatment was occasionally present on brain MRIs from prior SRS treatments. Most RIMs were + DC, and some subsequent SRS courses treated only + DC RIMs. These findings suggest enhanced BM detection might enable earlier treatment and reduce the need for additional SRS.

Artificial-intelligence-driven measurements of brain metastases'response to SRS compare favorably with current manual standards of assessment.

Evaluation of treatment response for brain metastases (BMs) following stereotactic radiosurgery (SRS) becomes complex as the number of treated BMs increases. This study uses artificial intelligence (AI) to track BMs after SRS and validates its output compared with manual measurements. Patients with BMs who received at least one course of SRS and followed up with MRI scans were retrospectively identified. A tool for automated detection, segmentation, and tracking of intracranial metastases on longitudinal imaging, MEtastasis Tracking with Repeated Observations (METRO), was applied to the dataset. The longest three-dimensional (3D) diameter identified with METRO was compared with manual measurements of maximum axial BM diameter, and their correlation was analyzed. Change in size of the measured BM identified with METRO after SRS treatment was used to classify BMs as responding, or not responding, to treatment, and its accuracy was determined relative to manual measurements. From 71 patients, 176 BMs were identified and measured with METRO and manual methods. Based on a one-to-one correlation analysis, the correlation coefficient was R2 = 0.76 (P = .0001). Using modified BM response classifications of BM change in size, the longest 3D diameter data identified with METRO had a sensitivity of 0.72 and a specificity of 0.95 in identifying lesions that responded to SRS, when using manual axial diameter measurements as the ground truth. Using AI to automatically measure and track BM volumes following SRS treatment, this study showed a strong correlation between AI-driven measurements and the current clinically used method: manual axial diameter measurements.

Automated Detection, Segmentation, and Tracking of Brain Metastases in Repeated Courses of Stereotactic Radiosurgery Using Integrated Artificial Intelligence.

Salvage stereotactic radiosurgery (SRS) for distant brain metastases has been demonstrated as a safe and effective approach for intracranial recurrences after initial SRS. However, accurate tumor detection and segmentation among responding tumors within the irradiated parenchyma can be challenging. The requirement for the registration and reference to the previous course of SRS is very time-consuming and suffers significant inter and intra-reader variability. Artificial intelligence (AI)-assisted system has been proven to improve the accuracy and efficiency in the clinical flow of de-novo SRS. We hypothesize that an integrated AI system can facilitate an automated tumor contouring process for repeated SRS. Three patients who underwent their third course of SRS to brain metastases were selected for the pioneering works. They have had two sessions of SRS with a mean lesion number of 4 and 3.7, respectively. VBrain, an FDA-approved brain tumor management AI platform, was used to co-registered serial MR scans and automatically identify, track, and contour brain metastases for each course of SRS. The AI also indicated new lesions and treated lesions for each course. Three radiation oncologists experienced in brain SRS contoured the gross tumor volumes (GTVs) of the third course of SRS in two reader modes (assisted then unassisted) with a memory washout period of one week between each section. The segmentation ground truth was established through consensus among the three experts. Lesion-wise sensitivity, contouring accuracy, and consuming time were compared between the two contouring modes. In each patient, there were 15, 11, and 9 metastases, with a median diameter of 4.72 (95% CI: 4.05, 6.91) mm. The mean lesion-wise sensitivity was 96.96±2.47% with AI assistance and 76.90 ± 7.10% without assistance. There were two false-positive lesions in the assisted read, resulting in a low average false-positive rate of 0.67 per patient, while no false positive for the unassisted mode. AI assistance improved contouring accuracy. The median Dice similarity coefficient (DSC) was 0.71 (95% CI: 0.55, 0.87) for assisted contouring and 0.65 (95% CI: 0.46, 0.85) for unassisted contouring. We also use average Hausdorff distance (HD) to measure segmentation results. The mean HD was 0.72± 0.13 mm versus 0.73±0.08 mm for the two contouring modes (p = 0.02) Furthermore, the median contouring time per case was significantly shorter with AI assistance than without assistance (20.8 minutes vs. 29.8 minutes; p < 0.001), corresponding to a 43.2% time-saving. Our results suggest that the integration of an AI-based system into repeated brain SRS can significantly improve the accuracy and efficiency of tumor detection and segmentation. This approach has the potential to streamline the treatment planning process for salvage SRS.

Impact of Concurrent Targeted Therapy and Immunotherapy on the Incidence of Radiation Necrosis Following Stereotactic Radiosurgery for Brain Metastases

The management of metastatic disease has evolved with the advent of immunotherapy (IT) and targeted therapies (TT). Yet, there is limited understanding of the toxicity associated with combining these agents with stereotactic radiosurgery (SRS). We retrospectively evaluated the impact of concurrent systemic therapy (ST) on the risk of radiation necrosis (RN) following LINAC-based SRS for brain metastases (BM). A retrospective study was conducted within an integrated health care system from March 2017 to December 2021 of 313 patients who underwent SRS or fractionated SRS in 3 or 5 treatments to a total of 1,644 intact BM. Post-operative cavity SRS and re-irradiated lesions were excluded. RN was diagnosed using perfusion MRI, contrast clearance MRI, or serial standard MRI and graded using CTCAE (v.5). Concurrent ST was defined as administration within 1 month preceding or following SRS. Overall survival (OS) and risk of RN were estimated by the Kaplan-Meier method. Logistics regression analyses were performed to compare risk of RN in patients who received concurrent systemic therapy to those who did not, adjusted for PTV volume and receipt of whole brain radiotherapy (WBRT). Median follow-up was 12.2 months. Median age was 64 years (range: 24-92). Primary sites per patient included lung (48.9%), breast (18.2%), melanoma (11.5%), kidney (6.1%), and other (15.3%). Median total lesions treated was 3 (range: 1-44); 65.9% of patients underwent 1 course of SRS, 23.4% underwent 2 courses, 6.2% underwent 3 courses, 4.5% underwent >4 courses. Seventy-six (24.2%) patients received WBRT. Overall, 70.6% of lesions received concurrent ST including chemotherapy (CT) (32.5%), IT (26.8%), and TT (27.6%); 16.4% received a combination of ST. Median OS was 12.9 months (95% CI: 10.4-15.5). RN was observed in 50 (3.0%) lesions in 42 (13.4%) patients. The 1-year risk of RN was 4.0% per lesion and 15.4% per patient. Symptomatic RN (SRN) was observed in 31 (1.9%) lesions in 24 (7.7%) patients. The 1-year risk of SRN was 2.7% per lesion and 10.1% per patient. When compared to lesions treated without concurrent systemic therapy, there was no increased risk of RN observed in lesions treated with concurrent CT (adjusted OR = 0.86 (95% CI: 0.43-1.73) p = 0.68), concurrent IT (adjusted OR = 0.84 (95% CI: 0.41-1.71) p = 0.84), or concurrent TT (adjusted OR = 0.57 (95% CI: 0.25-1.30) p = 0.18). Treatments of SRN included dexamethasone (96.8%), bevacizumab (22.6%), and laser interstitial thermal therapy (6.5%). Concurrent IT and TT appears well-tolerated in patients who undergo SRS for treatment of BM. No increased risk of RN was observed in lesions treated with concurrent IT or TT compared to lesions treated in the absence of concurrent ST. Further prospective and agent-specific evaluation is necessary to confirm these findings.

Intracranial Control with Combined Dual Immune-Checkpoint Blockade and SRS for Melanoma and NSCLC Brain Metastases

It is unknown whether the use of dual immune-checkpoint inhibition (D-ICI) combined with stereotactic radiosurgery (SRS) affects local control of brain metastases (BMs). We sought to characterize the efficacy of SRS and D-ICI in patients with BMs in a large, single-institution cohort. Patients with melanoma and non-small cell lung cancer (NSCLC) BMs treated with SRS from January 1, 2016 to August 1, 2022 were evaluated. Patients were stratified by treatment with D-ICI versus single ICI (S-ICI). Concurrent ICI was defined as ICI given within four weeks of SRS. Local recurrence (LR), intracranial progression (IP), and overall survival (OS) were estimated using competing risk and Kaplan-Meier analyses. IP included both local and distant intracranial recurrence. One thousand seven hundred four SRS-treated BMs from 288 patients met inclusion criteria. 55% of patients were symptomatic from their BMs at presentation. Median age, KPS, number of lesions, and SRS courses were 64 (Q1Q3:56-70.5), 90 (80-90), 2 (1-4), and 1 (1-2), respectively. One hundred twenty-eight (44%) melanoma and 160 (56%) NSCLC patients were included. 82 (28.5%), 129 (44.8%), and 77 (26.7%) patients were treated with D-ICI, S-ICI, or SRS alone. Median SRS dose, fractions, and PTV were 20 (Q1Q3:20-25), 1 (1-5), and 0.3cc3 (0.1-1.2). The median follow-up was 14.3 months. One hundred twenty-seven (7.45%) BMs recurred post-SRS and the median time to LR was 4.8 months (Q1Q3:3.0-9.2). On competing risk analysis, LR was significantly reduced with D-ICI (HR: 0.452, p = 0.0024), but not with S-ICI (HR: 0.693, p = 0.0596) compared to SRS alone. The 1-year LR was 3.77% (95% CI = 2.19-6.00), 6.8% (5.19-8.70), and 8.96% (6.48-11.93) with D-ICI, S-ICI, and SRS alone. The median time to IP was 4.1 months (Q1Q3 = 2.9-9.5). On competing risk analysis, IP was significantly reduced with D-ICI (HR = 0.638, p = 0.031), but not with S-ICI (HR = 0.756, p = 0.106) compared to SRS alone. 1-year IP was 40.05% (95% CI = 29.14-50.70), 51.86% (42.78-60.19), and 58.49% (46.30-68.84) with D-ICI, S-ICI, and SRS alone. Concurrent delivery of D-ICI and SRS significantly reduced IP (HR = 0.463, p = 0.0071), whereas other combinations of timing and ICI did not reach significance. Median OS was 11.9 months after SRS. On Kaplan Meier analysis, OS was significantly improved with D-ICI (HR = 0.616, 95% CI = 0.412-0.923, p = 0.019), but not with S-ICI (HR = 0.877, 95% CI = 0.633-1.217, p = 0.433) compared to SRS alone. Hospitalizations (p = 0.021) and immune-related adverse events (irAEs) (p<0.001) were increased with D-ICI. Any grade radiation necrosis (RN) was also increased with D-ICI (p = 0.013), but neurologic adverse events were comparable across cohorts (p = 0.572). D-ICI combined with SRS was associated with improved local control, intracranial control, and overall survival compared to SRS alone, whereas S-ICI was not associated with an improvement in these outcomes. However, D-ICI was also associated with increased risks of irAEs and RN.

CLRM-01 DETERMINANTS OF SYMPTOMATIC INTRACRANIAL PROGRESSION FOLLOWING AN INITIAL STEREOTACTIC RADIOSURGERY COURSE

Abstract To determine the prognostic significance of neurologic symptoms at progression for patients with brain metastases (BM) undergoing stereotactic radiosurgery (SRS), all patients completing an initial course of SRS between 1/2015 and 12/2020 were identified across two institutions. Intracranial progression (ICP) was recorded, with symptomatic ICP (SICP) defined as new/worsening neurologic symptoms without another etiology. Overall survival (OS), freedom from ICP (FFICP), and freedom from symptomatic ICP (FFSICP) were assessed via Kaplan-Meier method. For FFSICP, patients were censored at time of death or asymptomatic ICP. Logistic regression models tested parameter association to neurologic symptoms. Cox proportional hazard models tested parameters impacting FFICP and FFSICP. Among 1383 patients, median age was 63.4 years, 54.8% were female, and 45.6% had KPS ≥90. Common primary sites were non-small cell lung (48.7%), breast (14.7%), and melanoma (8.5%). 46.9% had oligometastatic disease (≤5 foci), and 53.4% had &amp;gt;1 BM. 26.1% patients had prior surgical resection, and 10.3% had WBRT. With a median follow up of 8.7 months, 504 (36%) and 194 (14%) experienced asymptomatic and symptomatic ICP respectively. OS was worse for patients experiencing SICP (median 10.2 vs 17.9 months, p&amp;lt;0.001). Among patients with ICP, SICP was associated with total treated tumor volume (per-ml, OR 1.01, 95% CI 1.00-1.02), while those with more recent MRI imaging and post-SRS systemic therapy was associated with asymptomatic ICP. Patients who received post-SRS chemotherapy (HR 0.64, 95% CI 0.45-0.90), immunotherapy (HR 0.49, 95% CI 0.34-0.71), or targeted therapy (HR 0.49, 95% CI 0.33-0.73) had better FFSCIP. Worse FFSCIP was associated with melanoma (HR 2.56, 95% CI 1.49-4.38), and greater than 2 BMs (HR 2.24, 95% CI 1.56-3.22). SICP is prognostic for OS, and is associated with melanoma, no systemic therapy post-SRS, and greater number of BMs. These data further support the use of SICP as a meaningful clinical endpoint.

LMAP-08 MULTI-INSTITUTIONAL CLINICAL OUTCOMES FOLLOWING STEREOTACTIC RADIOSURGERY FOR BRAIN METASTASES FROM GYNECOLOGIC MALIGNANCIES

Abstract Brain metastases (BM) are rare in patients with gynecologic (GYN) malignancies. We identified and analyzed all patients who completed a course of stereotactic radiosurgery (SRS) between 1/2015 and 12/2020 across two institutions. Demographic and clinical parameters were collected. Intracranial progression (ICP) was defined as any concern on post-SRS imaging for recurrence determined by multidisciplinary consensus. Clinical parameter distributions were compared across GYN and non-GYN patients utilizing chi-squared tests and t-tests. Overall survival (OS) and freedom from intracranial progression (FFICP) were estimated via the Kaplan Meier method. From a cohort of 1383 patients who completed SRS for BMs, 2.4% (n=33) had a BM from a GYN malignancy. Subsites included endometrial (n=19), ovarian (n=11), vulvar (n=2), cervical (n=1). There was no difference in age between GYN and non-GYN patients (mean 67.1 years vs. 63.0 years, p=0.06). GYN patients with BMs compared to non-GYN patients were more likely to have resection of BMs (48.5% vs. 25.6%, p=0.003), larger planned target volumes of all BMs (mean 23.6cc vs mean 15.2cc, p=0.02), and pre-SRS chemotherapy (72.7% vs. 49.6%, p=0.009). GYN patients were less likely to have whole brain radiation therapy prior to SRS (0 vs. 10.5%, p=0.049). GYN patients had worse OS compared to non-GYN patients (median 6.6 months (95% CI 3.3-12.4) vs. 10.0 months (95% CI 9.1-11.2), (HR 1.62 (95% CI 1.11-2.36), p=0.011)). Median FFICP for GYN patients was 9.5 months (95% CI 4.0-not reached) and for non-GYN patients was 8.7 months (95% CI 7.7-9.5). There was no difference in FFICP for GYN patients compared to non-GYN patients (HR 0.84 (95% CI 0.5-1.5), p=0.55). When comparing patients with BMs from endometrial cancer versus ovarian cancer, there was no difference in OS (p=0.13) or FFICP (p=0.58). These data may help guide treatment decisions and post-therapy surveillance in patients with BMs of GYN origin.

Effects of Ataxia-Telangiectasia Mutated Variants on Radionecrosis and Local Control After Stereotactic Radiation Surgery for Non-Small Cell Lung Cancer Brain Metastases

Genetic variants affecting the radiation response protein ataxia-telangiectasia mutated (ATM) have been associated with increased adverse effects of radiation but also with improved local control after conventional radiation therapy. However, it is unknown whether ATM variants affect rates of radionecrosis (RN) and local intracranial progression (LIP) after stereotactic radiosurgery (SRS) for brain metastases. Patients undergoing an initial course of SRS for non-small cell lung cancer (NSCLC) brain metastases at a single institution were retrospectively identified. Kaplan-Meier estimates were calculated and Cox proportional hazards testing was performed based on ATM variant status. A total of 541 patients completed SRS for brain metastasis secondary to NSCLC, of whom 260 completed molecular profiling. Variants of ATM were identified in 36 cases (13.8%). Among patients who completed molecular profiling, RN incidence was 4.9% (95% CI, 1.6%-8.2%) at 6 months and 9.9% (95% CI, 4.8%-15.0%) at 12 months. Incidence of RN was not significantly increased among patients with ATM variants, with an RN incidence of 5.3% (95% CI, 0.0%-15.3%) at both 6 and 12 months (P=.46). For all patients who completed genomic profiling, LIP was 5.4% (95% CI, 2.4%-8.4%) at 6 months and 9.8% (5.5%-14.1%) at 12 months. A significant improvement in LIP was not detected among patients with ATM variants, with an LIP incidence of 3.1% (0.0%-9.1%) at both 6 and 12 months (P=.26). Although differences according to ATM variant type (pathologic variant or variant of unknown significance) did not reach significance, no patients with ATM pathologic variants experienced LIP. We did not detect significant associations between ATM variant status and RN or LIP after SRS for NSCLC brain metastases. The current data set allows estimation of patient cohort sizes needed to power future investigations to identify genetic variants that associate with significant differences in outcomes after SRS.

Automatically tracking brain metastases after stereotactic radiosurgery.

Patients with brain metastases (BMs) are surviving longer and returning for multiple courses of stereotactic radiosurgery. BMs are monitored after radiation with follow-up magnetic resonance (MR) imaging every 2-3months. This study investigated whether it is possible to automatically track BMs on longitudinal imaging and quantify the tumor response after radiotherapy. The METRO process (MEtastasis Tracking with Repeated Observations was developed to automatically process patient data and track BMs. A longitudinal intrapatient registration method for T1 MR post-Gd was conceived and validated on 20 patients. Detections and volumetric measurements of BMs were obtained from a deep learning model. BM tracking was validated on 32 separate patients by comparing results with manual measurements of BM response and radiologists' assessments of new BMs. Linear regression and residual analysis were used to assess accuracy in determining tumor response and size change. A total of 123 irradiated BMs and 38 new BMs were successfully tracked. 66 irradiated BMs were visible on follow-up imaging 3-9months after radiotherapy. Comparing their longest diameter changes measured manually vs. METRO, the Pearson correlation coefficient was 0.88 (p<0.001); the mean residual error was -8±17%. The mean registration error was 1.5±0.2mm. Automatic, longitudinal tracking of BMs using deep learning methods is feasible. In particular, the software system METRO fulfills a need to automatically track and quantify volumetric changes of BMs prior to, and in response to, radiation therapy.

Intracranial and Extracranial Progression and Their Correlation With Overall Survival After Stereotactic Radiosurgery in a Multi-institutional Cohort With Brain Metastases

Clinical trials for metastatic malignant neoplasms are increasingly being extended to patients with brain metastases. Despite the preeminence of progression-free survival (PFS) as a primary oncologic end point, the correlation of intracranial progression (ICP) and extracranial progression (ECP) events with overall survival (OS) is poorly understood for patients with brain metastases following stereotactic radiosurgery (SRS). To determine the correlation of ICP and ECP with OS among patients with brain metastases completing an initial SRS course. This multi-institutional retrospective cohort study was conducted from January 1, 2015, to December 31, 2020. We included patients who completed an initial course of SRS for brain metastases during the study period, including receipt of single and/or multifraction SRS, prior whole-brain radiotherapy, and brain metastasis resection. Data analysis was performed on November 15, 2022. Non-OS end points included intracranial PFS, extracranial PFS, PFS, time to ICP, time to ECP, and any time to progression. Progression events were radiologically defined, incorporating multidisciplinary clinical consensus. The primary outcome was correlation of surrogate end points to OS. Clinical end points were estimated from time of SRS completion via the Kaplan-Meier method, while end-point correlation to OS was measured using normal scores rank correlation with the iterative multiple imputation approach. This study included 1383 patients, with a mean age of 63.1 years (range, 20.9-92.8 years) and a median follow-up of 8.72 months (IQR, 3.25-19.68 months). The majority of participants were White (1032 [75%]), and more than half (758 [55%]) were women. Common primary tumor sites included the lung (757 [55%]), breast (203 [15%]), and skin (melanoma; 100 [7%]). Intracranial progression was observed in 698 patients (50%), preceding 492 of 1000 observed deaths (49%). Extracranial progression was observed in 800 patients (58%), preceding 627 of 1000 observed deaths (63%). Irrespective of deaths, 482 patients (35%) experienced both ICP and ECP, 534 (39%) experienced ICP (216 [16%]) or ECP (318 [23%]), and 367 (27%) experienced neither. The median OS was 9.93 months (95% CI, 9.08-11.05 months). Intracranial PFS had the highest correlation with OS (ρ = 0.84 [95% CI, 0.82-0.85]; median, 4.39 months [95% CI, 4.02-4.92 months]). Time to ICP had the lowest correlation with OS (ρ = 0.42 [95% CI, 0.34-0.50]) and the longest median time to event (median, 8.76 months [95% CI, 7.70-9.48 months]). Across specific primary tumor types, correlations of intracranial PFS and extracranial PFS with OS were consistently high despite corresponding differences in median outcome durations. The results of this cohort study of patients with brain metastases completing SRS suggest that intracranial PFS, extracranial PFS, and PFS had the highest correlations with OS and time to ICP had the lowest correlation with OS. These data may inform future patient inclusion and end-point selection for clinical trials.

Hypofractionated stereotactic radiosurgery (HSRS) as a salvage treatment for brain metastases failing prior stereotactic radiosurgery (SRS)

Various treatment options exist to salvage stereotactic radiosurgery (SRS) failures for brain metastases, including repeat SRS and hypofractionated SRS (HSRS). Our objective was to report outcomes specific to salvage HSRS for brain metastases that failed prior HSRS/SRS. Patients treated with HSRS to salvage local failures (LF) following initial HSRS/SRS, between July 2010 and April 2020, were retrospectively reviewed. The primary outcomes were the rates of LF, radiation necrosis (RN), and symptomatic radiation necrosis (SRN). Univariable (UVA) and multivariable (MVA) analyses using competing risk regression were performed to identify predictive factors for each endpoint. 120 Metastases in 91 patients were identified. The median clinical follow up was 13.4 months (range 1.1-111.1), and the median interval between SRS courses was 13.1 months (range 3.0-56.5). 115 metastases were salvaged with 20-35Gy in 5 fractions and the remaining five with a total dose ranging from 20 to 24Gy in 3-fractions. 67 targets (56%) were postoperative cavities. The median re-treatment target volume and biological effective dose (BED10) was 9.5cc and 37.5Gy, respectively. The 6- and 12- month LF rates were 18.9% and 27.7%, for RN 13% and 15.6%, and for SRN were 6.1% and 7.0%, respectively. MVA identified larger re-irradiation volume (hazard ratio [HR] 1.02, p = 0.04) and shorter interval between radiosurgery courses (HR 0.93, p < 0.001) as predictors of LF. Treatment of an intact target was associated with a higher risk of RN (HR 2.29, p = 0.04). Salvage HSRS results in high local control rates and toxicity rates that compare favorably to thosesingle fraction SRS re-irradiationexperiences reported in the literature.

Upfront and Repeated Stereotactic Radiosurgery in Patients With Brain Metastases From NSCLC

IntroductionApproximately 40% of non-small-cell lung cancer (NSCLC) patients develop brain metastases (BM). Stereotactic radiosurgery (SRS) instead of whole-brain radiotherapy (WBRT) is increasingly administered as an upfront treatment to patients with a limited number of BM. We present outcomes and validation of prognostic scores for these patients treated with upfront SRS. MethodsWe retrospectively analyzed 199 patients with a total of 268 SRS courses for 539 brain metastases. Median patient age was 63 years. For larger BM, dose reduction to 18 Gy or hypofractionated SRS in 6 fractions was applied. We analyzed the BMV-, the RPA-, the GPA- and the lung-mol GPA score. Cox proportional hazards models with univariate and multivariate analyses were fitted for overall survival (OS) and intracranial progression-free survival (icPFS). ResultsSixty-four patients died, 7 of them of neurological causes. Thirty eight patients (19,3%) required a salvage WBRT. Median OS was 38, 8 months (IQR: 6-NA). In univariate analysis as well as multivariate analysis, the Karnofsky performance scale index (KPI) ≥90% (P = 0, 012 and P = 0, 041) remained as independent prognostic factor for longer OS. All 4 prognostic scoring indices could be validated for OS assessment (BMV P = 0, 007; RPA P = 0, 026; GPA P = 0, 003; lung-mol GPA P = 0, 05). ConclusionIn this large cohort of NSCLC patients with BM treated with upfront and repeated SRS, OS was markedly favourable, in comparison to literature. Upfront SRS is an effective treatment approach in those patients and can decidedly reduce the impact of BM on overall prognosis. Furthermore, the analysed scores are useful prognostic tools for OS prediction.

A Deep Learning-Based Computer Aided Detection (CAD) System for Difficult-to-Detect Brain Metastases

We sought to develop a computer-aided detection (CAD) system that optimally augments human performance, excelling especially at identifying small inconspicuous brain metastases (BMs), by training a convolutional neural network on a unique magnetic resonance imaging (MRI) data set containing subtle BMs that were not detected prospectively during routine clinical care. Patients receiving stereotactic radiosurgery (SRS) for BMs at our institution from 2016 to 2018 without prior brain-directed therapy or small cell histology were eligible. For patients who underwent 2 consecutive courses of SRS, treatment planning MRIs from their initial course were reviewed for radiographic evidence of an emerging metastasis at the same location as metastases treated in their second SRS course. If present, these previously unidentified lesions were contoured and categorized as retrospectively identified metastases (RIMs). RIMs were further subcategorized according to whether they did (+DC) or did not (-DC) meet diagnostic imaging-based criteria to definitively classify them as metastases based upon their appearance in the initial MRI alone. Prospectively identified metastases (PIMs) from these patients, and from patients who only underwent a single course of SRS, were also included. An open-source convolutional neural network architecture was adapted and trained to detect both RIMs and PIMs on thin-slice, contrast-enhanced, spoiled gradient echo MRIs. Patients were randomized into 5 groups: 4 for training/cross-validation and 1 for testing. One hundred thirty-five patients with 563 metastases, including 72 RIMS, met criteria. For the test group, CAD sensitivity was 94% for PIMs, 80% for +DC RIMs, and 79% for PIMs and +DC RIMs with diameter <3 mm, with a median of 2 false positives per patient and a Dice coefficient of 0.79. Our CAD model, trained on a novel data set and using a single common MR sequence, demonstrated high sensitivity and specificity overall, outperforming published CAD results for small metastases and RIMs - the lesion types most in need of human performance augmentation.

Early Imaging Characteristics Associated with Development of Future Brain Metastases in Patients Undergoing Stereotactic Radiosurgery

<h3>Purpose/Objective(s)</h3> Brain MRIs are carefully reviewed during the process of stereotactic radiosurgery (SRS) planning to select targets with radiographic characteristics typical of brain metastases (BM). Some BMs, especially ones that are small, dural-based, appear similar to blood vessels or resemble microinfarcts, are difficult to definitively identify and may go untreated initially, only to become apparent on future imaging. We hypothesized that in patients receiving multiple courses of SRS, retrospective review of the initial planning MRI would reveal early evidence of lesions that would subsequently develop into metastases requiring SRS. <h3>Materials/Methods</h3> Patients undergoing two or more courses of SRS to BMs within a 6-month interval between 2016 and 2018 were included in this single-institution, retrospective study. Exclusion criteria included small cell histology or previous whole-brain radiation therapy. Brain MRIs from the initial treatment course were reviewed for the presence of a contrast enhancing lesion at the same location as a lesion treated in a subsequent course of SRS, which was classified as a "retrospectively identified metastasis" or RIM. <h3>Results</h3> Of the 683 patients who underwent 923 courses of SRS within the study period, 98 patients met inclusion criteria. There were 120 repeat courses of SRS, with 345 treated metastases in the subsequent course, 128 of which were RIMs. 48% were within 5mm of the dura: 12% were abutting blood vessels: 8% became symptomatic. 23% of subsequent courses consisted of metastases that were all identified as RIMs. <h3>Conclusion</h3> Radiographic evidence of lesions requiring future treatment was occasionally present on brain MRIs from prior SRS treatments. RIMs are typically small, often abut blood vessels and/or the dura, and rarely become symptomatic. In some patients undergoing SRS, RIMs alone were treated, suggesting that enhanced detection of RIMs could reduce the need for additional courses of SRS.

Dedicated Isotropic 3-D T1 SPACE Sequence Imaging for Radiosurgery Planning Improves Brain Metastases Detection and Reduces the Risk of Intracranial Relapse

<h3>Purpose/Objective(s)</h3> Stereotactic radiosurgery (SRS) is increasingly used for patients with brain metastasis, yet the risk of distant intracranial failure (DIF) continues to be the principal disadvantage of this focal approach. The objective of this study was to determine if advanced dedicated treatment planning images at SRS would improve lesion detection and consequentially reduce DIF. <h3>Materials/Methods</h3> Patients treated with SRS at a single tertiary care institution underwent dedicated treatment planning MRIs including a standard three-dimensional (3D) magnetization-prepared rapid acquisition with gradient echo (MPRAGE) sequence as well as an optimized 3D sampling perfection with application-optimized contrasts using different flip angle evolutions (SPACE) sequence from 02/2020-01/2021. DIF was calculated using the Kaplan-Meier method and comparisons were made to a historical consecutive cohort of patients treated using MPRAGE alone (02/2019-01/2020). <h3>Results</h3> 134 patients underwent 170 SRS courses for 821 brain metastases imaged with both MPRAGE and SPACE (primary cohort). MPRAGE sequence evaluation alone detected 679 (82.7%) lesions. With neuroradiologists evaluating SPACE and MPRAGE, an additional 108 lesions were identified (p<0.001). Upon multidisciplinary review, 34 (4.1%) additional lesions were identified. Compared to the historical cohort (103 patients, 133 SRS courses, 479 brain metastases) treated using MPRAGE alone, patients treated using MPRAGE and SPACE had improved median time to DIF (13.5 vs. 5.1 months, <i>p</i>=0.004). The benefit was even more pronounced for patients treated for their first SRS course (18.4 vs. 6.3 months, <i>p</i>=0.001). On multivariate analysis, factors associated with improved DIF included stable extracranial disease, better performance status, using MPRAGE and SPACE images, and being treated for first course SRS (all <i>p</i><0.05). SRS using MPRAGE and SPACE was associated with a 60% reduction in risk of DIF compared with a historical cohort using MPRAGE alone (HR: 0.40; 95%CI: 0.28, 0.57, <i>p</i><0.001). <h3>Conclusion</h3> Among patients with brain metastasis treated with SRS, the addition of a dedicated treatment planning SPACE sequence to MPRAGE substantially improved detection of brain metastasis compared to MPRAGE alone. This approach was associated with a statistically significant and clinically meaningful prolongation in time to DIF, especially for patients undergoing their first course of SRS.

Prognostic Model for Intracranial Progression after Stereotactic Radiosurgery: A Multicenter Validation Study.

Stereotactic radiosurgery (SRS) is a standard of care for many patients with brain metastases. To optimize post-SRS surveillance, this study aimed to validate a previously published nomogram predicting post-SRS intracranial progression (IP). We identified consecutive patients completing an initial course of SRS across two institutions between July 2017 and December 2020. Patients were classified as low- or high-risk for post-SRS IP per a previously published nomogram. Overall survival (OS) and freedom from IP (FFIP) were assessed via the Kaplan−Meier method. Assessment of parameters impacting FFIP was performed with univariable and multivariable Cox proportional hazard models. Among 890 patients, median follow-up was 9.8 months (95% CI 9.1−11.2 months). In total, 47% had NSCLC primary tumors, and 47% had oligometastatic disease (defined as ≤5 metastastic foci) at the time of SRS. Per the IP nomogram, 53% of patients were deemed high-risk. For low- and high-risk patients, median FFIP was 13.9 months (95% CI 11.1−17.1 months) and 7.6 months (95% CI 6.4−9.3 months), respectively, and FFIP was superior in low-risk patients (p < 0.0001). This large multisite BM cohort supports the use of an IP nomogram as a quick and simple means of stratifying patients into low- and high-risk groups for post-SRS IP.

Impact of ATM Mutations on Brain Metastasis Control and Radionecrosis in Non-Small Cell Lung Cancer Patients Undergoing SRS

<h3>Purpose/Objective(s)</h3> While Ataxia Telangectasia Mutated gene (ATM) variants have been associated with improved local control after radiation, adverse radiation events have been reported following conventional radiotherapy for extracranial tumors. It remains unknown whether ATM variants affect rates of radiation necrosis (RN) and local intracranial progression (LIP) following stereotactic radiosurgery (SRS) for brain metastases. We questioned whether pathologic ATM variants increase rates of RN and/or decrease rates of LIP following SRS for brain metastasis in non-small cell lung cancer (NSCLC) patients. <h3>Materials/Methods</h3> Patients undergoing an initial course of SRS for NSCLC brain metastases between 1/2015 and 12/2020 were retrospectively identified at our institution. RN and LIP were determined by either biopsy or multi-disciplinary serial MRI review. Cumulative incidence of RN and LIP were estimated via the Kaplan Meier method. Cox proportional hazards testing was performed for variant classification (not detected [ND] vs variant of unknown significance [VUS] or pathogenic variant [PV]). A power analysis estimated that 34 events would need to be observed to detect a >4-fold hazard ratio for event. <h3>Results</h3> 541 patients completed SRS for brain metastasis secondary to NSCLC, of which 260 completed molecular profiling. Comprehensive Genomic Profiling (CGP) was performed on cell free circulating tumor DNA (61%), primary tumor (17%), extracranial metastasis (17%), and brain tumor (18%), including ≥2 sites for 33 (13%) patients. Availability of CGP data was significantly associated with more recent year of SRS completion (2018-20, 64% vs 2015-17, 25%; p<0.001), younger age at brain metastasis diagnosis (64.8 years [IQR 57.1-72.8] vs 67.8 years [IQR 59.6-72.8]; p=0.036), no prior tobacco use (22% vs 10%; p<0.001), adenocarcinoma histology (84% vs 73%; p=0.002), and metastatic disease at initial diagnosis (73% vs 63%; p=0.017). ATM variants were identified in 36 cases (13.8%; 13 PV, 23 VUS). For all patients with CPG, RN incidence was 4.9% (95% CI 1.6 – 8.2%) at 6 months and 9.9% (95% CI 4.8 - 15.0%) at 12 months and LIP was 5.4% (95% CI 2.4 – 8.4%) at 6 months and 9.8% (5.5 – 14.1%) at 12 months. For patients with ATM variants, RN incidence was 5.3% (0.0 - 15.3%) at both 6 and 12 months, with 2 total RN events at 5.7 months and 13.9 months following SRS, respectively; LIP was 3.1% (0.0 - 9.1%) at both 6 and 12 months, with one total LIP event at 2.0 months following SRS (P=0.46 and P=0.26 for RN and LIP, respectively). <h3>Conclusion</h3> We did not detect significant differences in RN or LIP risk following SRS for NSCLC brain metastases associated with ATM variants. These results suggest that at this time, ATM status should not inform SRS use when otherwise indicated for NSCLC brain metastases. This dataset additionally allowed us to estimate the number of patients needed to detect a given magnitude of relative risk for RN or LIP associated with a given variant. These data can inform future studies as CGP data accumulate.

Risk of Late or Recurrent Radiation Necrosis beyond 5 Years from Stereotactic Radiosurgery

<h3>Purpose/Objective(s)</h3> Radiation necrosis (RN) is a well-recognized complication most commonly occurring within 1-3 years of stereotactic radiosurgery (SRS). However, the incidence of late RN after this period is poorly described. This study characterizes the incidence of and risk factors for late or recurrent RN occurring five years after SRS. <h3>Materials/Methods</h3> This single-institution, retrospective analysis included patients treated with LINAC-based SRS (any indication) who were followed for over five years clinically and with serial MRI imaging. Late or recurrent RN was defined as any imaging changes with corresponding neurologic symptoms or deficits occurring after five years from treatment not attributable to disease recurrence. Associations with late RN were analyzed using Pearson coefficients, Mann-Whitney U tests, and Fisher's exact test. Univariate and multivariate analyses were performed using Cox proportional hazards model. <h3>Results</h3> We identified 219 patients who were treated to a cumulative 297 lesions over 246 SRS courses to a median dose of 17.0 Gy (range: 12-35 Gy). 290 (97.3%) were treated in a single fraction with 64 (21.5%) of the lesions having been adjuvantly following resection. The median follow-up time was 7.4 years (range: 5.1-16.1 years), and late or recurrent RN occurred in 19 (12.9%) patients and in 23 (7.7%) lesions at a median 6.1 years (range: 5.1-13.9 years) from SRS. 15 of the 23 (65.2%) lesions with late RN were managed successfully with steroids, bevacizumab, and/or anti-epileptic drugs. The remaining 8 (34.8%) were refractory to medical therapy and required resection, and histopathology confirmed necrosis without disease recurrence in each. Late or recurrent RN was more common among brain metastases and arterial-venous malformations or fistulas (Table 1). The mean dose in lesions that developed late RN was higher than for lesions that did not (18.5 ±1.5 Gy vs. 16.5± 4.1 Gy; p<0.001). Gross tumor volume >1.5cc (HR: 4.6, 95% CI: 1.8-11.6; p=0.001), PTV margins >1mm (HR: 3.1; 95%CI: 1.1-9.0, p=0.040), brain V12 Gy >5cc (HR: 18.7, 95% CI: 5.5-63.2; p<0.001), and a history or early, resolved RN (HR: 13.7, 95% CI: 5.1-36.8; p<0.001) were all associated with late or recurrent RN. On multivariate analysis, only brain V12 Gy >5cc (HR: 4.6 95% CI: 1.0-20.5; p=0.046) and a history of early, resolved RN (HR: 4.0; 95% CI: 1.2-12.7; p=0.021) remained significantly associated with late or recurrent RN. <h3>Conclusion</h3> This is the first study, to our knowledge, describing the risk of late or recurrent RN beyond five years from SRS. While uncommon, late or recurrent RN was associated with brain V12 Gy >5 cc a history of early RN, suggesting that closer long-term follow-up may be warranted for these select patients.