Intricacies of Human-AI Interaction in Dynamic Decision-Making for Precision Oncology: A Case Study in Response-Adaptive Radiotherapy.

Background: Adaptive treatment strategies that can dynamically react to individual cancer progression can provide effective personalized care. Longitudinal multi-omics information, paired with an artificially intelligent clinical decision support system (AI-CDSS) can assist clinicians in determining optimal therapeutic options and treatment adaptations. However, AI-CDSS is not perfectly accurate, as such, clinicians' over/under reliance on AI may lead to unintended consequences, ultimately failing to develop optimal strategies. To investigate such collaborative decision-making process, we conducted a Human-AI interaction case study on response-adaptive radiotherapy (RT). Methods: We designed and conducted a two-phase study for two disease sites and two treatment modalities-adaptive RT for non-small cell lung cancer (NSCLC) and adaptive stereotactic body RT for hepatocellular carcinoma (HCC)-in which clinicians were asked to consider mid-treatment modification of the dose per fraction for a number of retrospective cancer patients without AI-support (Unassisted Phase) and with AI-assistance (AI-assisted Phase). The AI-CDSS graphically presented trade-offs in tumor control and the likelihood of toxicity to organs at risk, provided an optimal recommendation, and associated model uncertainties. In addition, we asked for clinicians' decision confidence level and trust level in individual AI recommendations and encouraged them to provide written remarks. We enrolled 13 evaluators (radiation oncology physicians and residents) from two medical institutions located in two different states, out of which, 4 evaluators volunteered in both NSCLC and HCC studies, resulting in a total of 17 completed evaluations (9 NSCLC, and 8 HCC). To limit the evaluation time to under an hour, we selected 8 treated patients for NSCLC and 9 for HCC, resulting in a total of 144 sets of evaluations (72 from NSCLC and 72 from HCC). Evaluation for each patient consisted of 8 required inputs and 2 optional remarks, resulting in up to a total of 1440 data points. Results: AI-assistance did not homogeneously influence all experts and clinical decisions. From NSCLC cohort, 41 (57%) decisions and from HCC cohort, 34 (47%) decisions were adjusted after AI assistance. Two evaluations (12%) from the NSCLC cohort had zero decision adjustments, while the remaining 15 (88%) evaluations resulted in at least two decision adjustments. Decision adjustment level positively correlated with dissimilarity in decision-making with AI [NSCLC: ρ = 0.53 ( p < 0.001); HCC: ρ = 0.60 ( p < 0.001)] indicating that evaluators adjusted their decision closer towards AI recommendation. Agreement with AI-recommendation positively correlated with AI Trust Level [NSCLC: ρ = 0.59 ( p < 0.001); HCC: ρ = 0.7 ( p < 0.001)] indicating that evaluators followed AI's recommendation if they agreed with that recommendation. The correlation between decision confidence changes and decision adjustment level showed an opposite trend [NSCLC: ρ = -0.24 ( p = 0.045), HCC: ρ = 0.28 ( p = 0.017)] reflecting the difference in behavior due to underlying differences in disease type and treatment modality. Decision confidence positively correlated with the closeness of decisions to the standard of care (NSCLC: 2 Gy/fx; HCC: 10 Gy/fx) indicating that evaluators were generally more confident in prescribing dose fractionations more similar to those used in standard clinical practice. Inter-evaluator agreement increased with AI-assistance indicating that AI-assistance can decrease inter-physician variability. The majority of decisions were adjusted to achieve higher tumor control in NSCLC and lower normal tissue complications in HCC. Analysis of evaluators' remarks indicated concerns for organs at risk and RT outcome estimates as important decision-making factors. Conclusions: Human-AI interaction depends on the complex interrelationship between expert's prior knowledge and preferences, patient's state, disease site, treatment modality, model transparency, and AI's learned behavior and biases. The collaborative decision-making process can be summarized as follows: (i) some clinicians may not believe in an AI system, completely disregarding its recommendation, (ii) some clinicians may believe in the AI system but will critically analyze its recommendations on a case-by-case basis; (iii) when a clinician finds that the AI recommendation indicates the possibility for better outcomes they will adjust their decisions accordingly; and (iv) When a clinician finds that the AI recommendation indicate a worse possible outcome they will disregard it and seek their own alternative approach.

Treatment of Central Nervous System Tumors on Combination MR-Linear Accelerators: Review of Current Practice and Future Directions.

Magnetic resonance imaging (MRI) provides excellent visualization of central nervous system (CNS) tumors due to its superior soft tissue contrast. Magnetic resonance-guided radiotherapy (MRgRT) has historically been limited to use in the initial treatment planning stage due to cost and feasibility. MRI-guided linear accelerators (MRLs) allow clinicians to visualize tumors and organs at risk (OARs) directly before and during treatment, a process known as online MRgRT. This novel system permits adaptive treatment planning based on anatomical changes to ensure accurate dose delivery to the tumor while minimizing unnecessary toxicity to healthy tissue. These advancements are critical to treatment adaptation in the brain and spinal cord, where both preliminary MRI and daily CT guidance have typically had limited benefit. In this narrative review, we investigate the application of online MRgRT in the treatment of various CNS malignancies and any relevant ongoing clinical trials. Imaging of glioblastoma patients has shown significant changes in the gross tumor volume over a standard course of chemoradiotherapy. The use of adaptive online MRgRT in these patients demonstrated reduced target volumes with cavity shrinkage and a resulting reduction in radiation dose to uninvolved tissue. Dosimetric feasibility studies have shown MRL-guided stereotactic radiotherapy (SRT) for intracranial and spine tumors to have potential dosimetric advantages and reduced morbidity compared with conventional linear accelerators. Similarly, dosimetric feasibility studies have shown promise in hippocampal avoidance whole brain radiotherapy (HA-WBRT). Next, we explore the potential of MRL-based multiparametric MRI (mpMRI) and genomically informed radiotherapy to treat CNS disease with cutting-edge precision. Lastly, we explore the challenges of treating CNS malignancies and special limitations MRL systems face.

Characterization of Immune Infiltrates Associated With Radiation Necrosis in the Setting of Brain Metastases Following Stereotactic Radiosurgery and Immunotherapy: A Retrospective Cohort Analysis.

INTRODUCTION: Radiation necrosis (RN) is caused by vascular damage and brain parenchymal injury resulting in inflammation following radiotherapy (RT) for brain metastases. The impact of immunotherapy (IO) on the immune cellular microenvironment in patients' brain metastases is unknown. The objective of this study was to characterize the inflammatory microenvironment in the setting of RN compared to recurrent metastasis and determine whether IO treatment affects the cellular infiltrates. METHODS: Adult patients with brain metastases from solid tumors who received both systemic IO and RT prior to resection of intracranial lesions were retrospectively analyzed. The resection either showed biopsy-proven RN or recurrent tumor. A group of patients who developed RN and were not on IO was reviewed as well. A total of 18 patients were categorized into one of three groups: necrosis, IO+RT; tumor, IO+RT; and necrosis, RT. Surgical specimens were stained for immune and inflammatory components and reviewed by a neuro-pathologist who remained blinded during the analysis. The presence or absence of lymphocytes, perivascular cuffs, plasma cells, macrophages, and fibrinoid vascular changes was characterized in a semiquantitative manner. RESULTS: The median age was 61.5 years (range 37-82 years). Seventy-seven percent of primary cancers were melanoma. Patients with RN were more likely to exhibit immune infiltrates compared to patients with recurrent metastasis. Limited analysis showed 100% of patients in "necrosis, IO+RT" had quantifiable cell counts; conversely, 83.3% of patients in "tumor, IO+RT" lacked quantifiable cell counts. Additionally, 83.3% of patients in "necrosis, RT" showed immune cells, including lymphocytes, macrophages, plasma cells, and cuffing. CONCLUSION: The immune microenvironment of brain metastasis following RT and IO showed higher levels of cell infiltrates in the RN setting versus the recurrent tumor setting. Patients who received prior IO compared to those with no IO had similar immune cell infiltrates adjacent to RN. Lower levels of immune cells in tumor recurrence following IO and RT raise the possibility that an environment lacking primed immune cells may decrease the efficacy of IO.

Palliative whole brain radiation therapy: an international state of practice.

BACKGROUND: Improvements in radiation delivery and systemic therapies have resulted in few remaining indications for palliative whole brain radiation therapy (WBRT). Most centers preferentially use stereotactic radiotherapy (SRT) and reserve WBRT for those with >15 lesions, leptomeningeal presentation, rapidly progressive disease, or limited estimated survival. Despite regional differences among preferred dose, fractionation, and treatment technique, we predict survival post-WBRT will remain poor-indicating appropriate application of WBRT in this era of SRT and improved systemic therapies. METHODS: A multi-center, international retrospective analysis of patients receiving WBRT in 2022 was performed. Primary end point was survival after WBRT. De-identified data were analyzed centrally. Patients receiving WBRT as part of a curative regimen, prophylactically, or as bridging therapy were excluded. The collected data consisted of patient parameters including prescription dose and fractionation, use of neurocognitive sparing techniques and survival after WBRT. Survival was calculated via the Kaplan-Meier method. RESULTS: Of 29,943 international RT prescriptions written at ten participating centers in 2022, 462 (1.5%) were for palliative WBRT. Participating centers were in the United States (n=138), the United Kingdom (n=111), Hong Kong (n=72), Italy (n=49), Belgium (n=45), Germany (n=27), Ghana (n=15), and Cyprus (n=5). Twenty-six different dose regimens were used. The most common prescriptions were for 3,000 cGy over 10 fractions (45.0%) and 2,000 cGy over 5 fractions (43.5%) with significant regional preferences (P<0.001). Prior SRT was delivered in 32 patients (6.7%), hippocampal avoidance (HA) was used in 44 patients (9.5%), and memantine was prescribed in 93 patients (20.1%). Survival ranged from 0 days to still surviving at 402 days post-treatment. The global median overall survival (OS) was 84 days after WBRT [95% confidence interval (CI): 68.0-104.0]. Actuarial survival at 7 days, 1 month, 3 months, and 6 months were 95%, 78%, 48%, and 32%, respectively. Twenty-seven patients (5.8%) were unable to complete their prescribed WBRT. CONCLUSIONS: This moment-in-time analysis confirms that patients with poor expected survival are being appropriately selected for WBRT-illustrating the dwindling indications for WBRT-and demonstrates the variance in global practice. Since poor survival precludes patients from deriving benefit, memantine and HA are best suited in carefully selected cases.

Stereotactic radiosurgery and anti-PD-1 + CTLA-4 therapy, anti-PD-1 therapy, anti-CTLA-4 therapy, BRAF/MEK inhibitors, BRAF inhibitors, or conventional chemotherapy for the management of melanoma brain metastases.

BACKGROUND: Immunotherapy and targeted BRAF/MEK inhibitors (i) have revolutionised the systemic management of advanced melanoma. Given the role of stereotactic radiosurgery (SRS) in the local management of brain metastases, we sought to evaluate clinical outcomes in patients with melanoma brain metastases (MBM) treated with SRS and various systemic therapies. METHODS: Patients were included if MBM were diagnosed and treated with SRS within 3 months of receiving anti-PD-1+CTLA-4 therapy, anti-PD-1 therapy, anti-CTLA-4 therapy, BRAF/MEK-i, BRAF-i, or conventional chemotherapy. Comparisons between groups were made for overall survival (OS), distant MBM control, local MBM, systemic progression-free survival (sPFS), and neurotoxicity. RESULTS: In total, 257 patients with 1048 MBM treated over 368 SRS sessions between 2011 and 2020 were identified. On MVA, treatment with anti-PD1+anti-CTLA-4, anti-PD-1, and BRAF/MEK-i improved distant intracranial control over conventional chemotherapy. No significant differences were noted in local control (LC) between groups (p = 0.78). Kaplan-Meier OS at 12 months for anti-PD-1 + CTLA-4 therapy, anti-PD-1 therapy, anti-CTLA-4 therapy, BRAF/MEK-i, BRAF-i, and conventional chemotherapy was 68%, 59%, 45%, 62%, 21%, and 15%, respectively (p = <0.0001). The sPFS rates at 12 months were 57%, 53%, 42%, 45%, 14%, and 6% (p = <0.0001). No significant differences were noted in rates of radiation necrosis (p = 0.93). CONCLUSIONS: This is among the largest series evaluating MBM treated with SRS and various systemic therapy regimens. Our analysis noted significant differences in OS, distant MBM control, and sPFS by systemic therapy. No differences in LC or radiation necrosis risk were noted.

Tucatinib and stereotactic radiosurgery in the management of HER2 positive breast cancer brain metastases.

PURPOSE: HER2-positive breast cancer has a high risk of brain metastasis. Stereotactic radiosurgery (SRS) is standard of care for limited brain metastases. Tucatinib, a HER2-targeted tyrosine kinase inhibitor, has demonstrated intracranial efficacy in the HER2-CLIMB Trial. However, it is unknown whether tucatinib with SRS is safe or effective. METHODS: A retrospective analysis of HER2-positive breast cancer treated with SRS and tucatinib for brain metastases management was performed. All patients received tucatinib and SRS for the management of active brain metastases. The primary endpoint was local and distant brain tumor control. Secondary endpoints were intracranial progression free survival (CNS-PFS), systemic PFS, overall survival (OS), and neurotoxicity. RESULTS: A total of 135 lesions treated with SRS over 39 treatment sessions in 22 patients were identified. Median follow-up from tucatinib initiation was 20.8 months. Local brain control was 94% at 12-months and 81% at 24-months. Distant brain control was 39% at 12-months and 26% at 24-months. Median survival was 21.2 months, with 12- and 24-month OS rates of 84% and 50%, respectively. Median CNS-PFS was 11.3 months, with 12- and 24-month CNS-PFS rates of 44.9% at both time points. Median systemic PFS was not reached, with 12- and 24-month systemic PFS rates of 86% and 57%, respectively. Symptomatic radiation necrosis occurred in 6 (4%) lesions. No additional unexpected toxicities were noted. CONCLUSIONS: SRS in combination with tucatinib, capecitabine, and trastuzumab appears to be a safe and feasible treatment for HER2 + brain metastases. Further prospective evaluation of potential synergistic effects is warranted.

Revisiting the Concept of Recurrence of Primary Central Nervous System Lymphomas After Complete Response to Methotrexate-Based Therapy: Periventricular Reseeding as the Predominant Mechanism of Recurrence.

Purpose: Understanding patterns of relapse for primary central nervous system lymphoma (PCNSL) may inform mechanisms of recurrence and optimal consolidation strategies. In this study, we report patterns of relapse among patients with PCNSL who achieved a complete response to high-dose methotrexate (HD-MTX)-based chemotherapy with or without consolidation radiation therapy (RT). Methods and Materials: We conducted an institutional retrospective analysis of patients with PCNSL who received HD-MTX-based chemotherapy between November 2001 and May 2019. Relapses were characterized as in-field (within original T1 contrasted lesion), marginal (within T2 fluid-attenuated inversion recovery but not T1), local (in-field or marginal), distant brain (no overlap), or distant (distant brain, cerebrospinal fluid, vitreous or extra-axial) and further characterized with respect to periventricular location (≤10 mm of ventricles). Results: Seventy-eight patients with PCNSL met inclusion criteria, of whom 29 (37%) underwent consolidation RT. Median progression-free survival and overall survival were 57.0 and 66.7 months, respectively. After a median follow-up of 38.9 months, a total of 32 patients (41%) experienced recurrence. Most patients (21 [65.6%]) had a periventricular failure. Surprisingly, local recurrences (n = 11) were exclusively observed within periventricular lesions, whereas distant recurrences (n = 21) were seen in both periventricular and nonperiventricular locations (P = .009). The median time to progression was shorter for locally recurrent lesions compared with distant recurrences (13.8 vs 26.1 months; P = .03). Conclusions: After complete response to HD-MTX, few failures occurred within initial T1 contrast-enhancing lesions and many of these may have been alternatively classified as periventricular failures. These observations argue against the use of purely focal RT consolidation for patients who achieve a complete response after HD-MTX-based chemotherapy and suggest that periventricular reseeding may have a central role in PCNSL recurrence.

Dosimetric feasibility of hippocampal avoidance whole brain radiotherapy with an MRI-guided linear accelerator.

PURPOSE/OBJECTIVE(S): Whole brain radiotherapy with hippocampal avoidance (HA-WBRT) is a technique utilized to treat metastatic brain disease while preserving memory and neurocognitive function. We hypothesized that the treatment planning and delivery of HA-WBRT plans is feasible with an MRI-guided linear accelerator (linac) and compared plan results with clinical non-MRI-guided C-Arm linac plans. MATERIALS/METHODS: Twelve HA-WBRT patients treated on a non-MRI-guided C-Arm linac were selected for retrospective analysis. Treatment plans were developed using a 0.35T MRI-guided linac system for comparison to clinical plans. Treatment planning goals were defined as provided in the Phase II Trial NRG CC001. MRI-guided radiotherapy (MRgRT) treatment plans were developed by a dosimetrist and compared with clinical plans. quality assurance (QA) plans were generated and delivered on the MRI-guided linac to a cylindrical diode detector array. Planning target volume (PTV) coverage was normalized to ∼95% to provide a control point for comparison of dose to the organs at risk. RESULTS: MRgRT plans were deliverable and met all clinical goals. Mean values demonstrated that the clinical plans were less heterogeneous than MRgRT plans with mean PTV V37.5 Gy of 0.00% and 0.03% (p = 0.013), respectively. Average hippocampi maximum doses were 14.19 ± 1.29 Gy and 15.00 ± 1.51 Gy, respectively. The gamma analysis comparing planned and measured doses resulted in a mean of 99.9% ± 0.12% of passing points (3%/2mm criteria). MRgRT plans had an average of 38.33 beams with average total delivery time and beam-on time of 13.7 (11.2-17.5) min and 4.1 (3.2-5.4) min, respectively. Clinical plan delivery times ranged from 3 to 7 min depending on the number of noncoplanar arcs. Planning time between the clinical and MRgRT plans was comparable. CONCLUSION: This study demonstrates that HA-WBRT can be treated using an MRI-guided linear accelerator with comparable treatment plan quality and delivery accuracy.

Improving Brain Metastases Outcomes Through Therapeutic Synergy Between Stereotactic Radiosurgery and Targeted Cancer Therapies.

Brain metastases are the most common form of brain cancer. Increasing knowledge of primary tumor biology, actionable molecular targets and continued improvements in systemic and radiotherapy regimens have helped improve survival but necessitate multidisciplinary collaboration between neurosurgical, medical and radiation oncologists. In this review, we will discuss the advances of targeted therapies to date and discuss findings of studies investigating the synergy between these therapies and stereotactic radiosurgery for non-small cell lung cancer, breast cancer, melanoma, and renal cell carcinoma brain metastases.

The presentation of brain metastases in melanoma, non-small cell lung cancer, and breast cancer and potential implications for screening brain MRIs.

PURPOSE: This study assessed the presentation and institutional outcomes treating brain metastases (BM) of breast cancer (BC), non-small cell lung cancer (NSCLC), and melanoma origin. METHODS: Patients with brain metastases treated between 2014 and 2019 with primary melanoma, NSCLC, and BC were identified. Overall survival (OS) was calculated from dates of initial BM diagnosis using the Kaplan-Meier method. RESULTS: A total of 959 patients were identified including melanoma (31%), NSCLC (51%), and BC (18%). Patients with BC were younger at BM diagnosis (median age: 57) than NSCLC (65) and melanoma patients (62, p < 0.0001). Breast cancer patients were more likely to present with at least 5 BM (27%) than NSCLC (14%) and melanoma (13%), leptomeningeal disease (23%, 6%, and 6%, p = 0.0004) and receive whole brain radiation therapy (WBRT) (58%, 37%, and 22%, p < 0.0001). There were no differences in surgical resection (24%, 24%, and 29%, p = 0.166). Median OS was shorter for BC patients (9.9, 10.3, and 13.7 months, p = 0.0006). CONCLUSION: Breast cancer patients were more likely to be younger, present with advanced disease, require WBRT, and have poorer OS than NSCLC and melanoma patients. Further investigation is needed to determine which BC patients are at sufficient risk for brain MRI screening.

Clinical outcomes of non-small cell lung cancer brain metastases treated with stereotactic radiosurgery and immune checkpoint inhibitors, EGFR tyrosine kinase inhibitors, chemotherapy and immune checkpoint inhibitors, or chemotherapy alone.

OBJECTIVE: Immune checkpoint inhibitors (ICIs) and epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKIs) are commonly used in the systemic management of non-small cell lung cancer (NSCLC) brain metastases (BMs). However, optimizing control of NSCLC BM with stereotactic radiosurgery (SRS) and various systemic therapies remains an area of investigation. METHODS: Between 2016 and 2019, the authors identified 171 NSCLC BM patients with 646 BMs treated with single-fraction SRS within 3 months of receiving treatment with ICIs (n = 56; 33%), EGFR-TKI (n = 30; 18%), chemotherapy and ICIs (n = 23; 14%), or standard chemotherapy alone (n = 62; 36%). Time-to-event analysis was conducted, and outcomes included distant intracranial control (DIC), local control (LC), and overall survival from SRS. RESULTS: The median follow-up from BM diagnosis was 8.9 months (range 0.3-127 months). The 12-month Kaplan-Meier DIC rates were 37%, 53%, 41%, and 21% (p = 0.047) for the ICI, EGFR-TKI, ICI and chemotherapy, and chemotherapy-alone groups, respectively. On multivariate analysis, DIC was improved with EGFR-TKI (HR 0.4, 95% CI 0.3-0.8, p = 0.005) compared with conventional chemotherapy and treatment with SRS before systemic therapy (HR 0.5, 95% CI 0.3-0.9, p = 0.03) compared with after; and LC was improved with SRS before (HR 0.4, 95% CI 0.2-0.9, p = 0.03) or concurrently (HR 0.3, 95% CI 0.1-0.6, p = 0.003) compared with after. No differences in radionecrosis were noted by timing or type of systemic therapy. CONCLUSIONS: The authors' analysis showed significant differences in DIC based on receipt of systemic therapy and treatment with SRS before systemic therapy improved DIC. Prospective evaluation of the potential synergism between systemic therapy and SRS in NSCLC BM management is warranted.

Long-Term Report of a Comprehensive Molecular and Genomic Analysis in NRG Oncology/RTOG 0424: A Phase II Study of Radiation and Temozolomide in High-Risk Grade II Glioma.

PURPOSE: This study sought to determine the prognostic significance of the WHO-defined glioma molecular subgroups along with additional alterations, including MGMT promoter methylation and mutations in ATRX, CIC, FUBP1, TERT, and TP53, in NRG/RTOG 0424 using long-term follow-up data. METHODS: Mutations were determined using an Ion Torrent sequencing panel. 1p/19q co-deletion and MGMT promoter methylation were determined by Affymetrix OncoScan and Illumina 450K arrays. Progression-free survival (PFS) and overall survival (OS) were estimated using the Kaplan-Meier method and tested using the log-rank test. Hazard ratios were calculated using the Cox proportional hazard model. Multivariable analyses (MVAs) included patient pretreatment characteristics. RESULTS: We obtained complete molecular data to categorize 80/129 eligible patients within the WHO subgroups. Of these, 26 (32.5%) were IDHmutant/co-deleted, 28 (35%) were IDHmutant/non-co-deleted, and 26 (32.5%) were IDHwild-type. Upon single-marker MVA, both IDHmutant subgroups were associated with significantly better OS and PFS (P values < .001), compared with the IDHwild-type subgroup. MGMT promoter methylation was obtained on 76 patients, where 58 (76%) were methylated and 18 (24%) were unmethylated. Single-marker MVAs demonstrated that MGMT promoter methylation was statistically significant for OS (P value < .001) and PFS (P value = .003). In a multimarker MVA, one WHO subgroup comparison (IDHmutant/co-deleted v IDHwild-type) was significant for OS (P value = .045), whereas MGMT methylation did not retain significance. CONCLUSION: This study reports the long-term prognostic effect of the WHO molecular subgroups, MGMT promoter methylation, and other mutations in NRG/RTOG 0424. These results demonstrate that the WHO molecular classification and MGMT both serve as strong prognostic indicators, but that MGMT does not appear to add statistically significant prognostic value to the WHO subgrouping, above and beyond IDH and 1p/19q status.

Capecitabine and stereotactic radiation in the management of breast cancer brain metastases.

BACKGROUND: Little is known about the safety and efficacy of concurrent capecitabine and stereotactic radiotherapy in the setting of breast cancer brain metastases (BCBM). METHODS: Twenty-three patients with BCBM underwent 31 stereotactic sessions to 90 lesions from 2005 to 2019 with receipt of capecitabine. The Kaplan-Meier method was used to calculate overall survival (OS), local control (LC), and distant intracranial control (DIC) from the date of stereotactic radiation. Imaging was independently reviewed by a neuro-radiologist. RESULTS: Median follow-up from stereotactic radiation was 9.2 months. Receptor types of patients treated included triple negative (n = 7), hormone receptor (HR)+/HER2- (n = 7), HR+/HER2+ (n = 6), and HR-/HER2+ (n = 3). Fourteen patients had stage IV disease prior to BCBM diagnosis. The median number of brain metastases treated per patient was 3 (1 to 12). The median dose of stereotactic radiosurgery (SRS) was 21 Gy (range: 15-24 Gy) treated in a single fraction and for lesions treated with fractionated stereotactic radiation therapy (FSRT) 25 Gy (24-30 Gy) in a median of 5 fractions (range: 3-5). Of the 31 stereotactic sessions, 71% occurred within 1 month of capecitabine. No increased toxicity was noted in our series with no cases of radionecrosis. The 1-year OS, LC, and DIC were 46, 88, and 30%, respectively. CONCLUSIONS: In our single institution experience, we demonstrate stereotactic radiation and capecitabine to be a safe treatment for patients with BCBM with adequate LC. Further study is needed to determine the potential synergy between stereotactic radiation and capecitabine in the management of BCBM.

Breast cancer subtype predicts clinical outcomes after stereotactic radiation for brain metastases.

PURPOSE: We investigated the prognostic ability of tumor subtype for patients with breast cancer brain metastases (BCBM) treated with stereotactic radiation (SRT). METHODS: This is a retrospective review of 181 patients who underwent SRT to 664 BCBM from 2004 to 2019. Patients were stratified by subtype: hormone receptor (HR)-positive, HER2-negative (HR+/HER2-), HR-positive, HER2-positive (HR+/HER2+), HR-negative, HER2-positive (HR-/HER2+), and triple negative (TN). The Kaplan-Meier method was used to calculate overall survival (OS), local control (LC), and distant intracranial control (DIC) from the date of SRT. Multivariate analysis (MVA) was conducted using the Cox proportional hazards model. RESULTS: Median follow up from SRT was 11.4 months. Of the 181 patients, 47 (26%) were HR+/HER2+, 30 (17%) were HR-/HER2+, 60 (33%) were HR+/HER2-, and 44 (24%) were TN. Of the 664 BCBMs, 534 (80%) received single fraction stereotactic radiosurgery (SRS) with a median dose of 21 Gy (range 12-24 Gy), and 130 (20%) received fractionated stereotactic radiation therapy (FSRT), with a median dose of 25 Gy (range 12.5-35 Gy) delivered in 3 to 5 fractions. One-year LC was 90%. Two-year DIC was 35%, 23%, 27%, and 16% (log rank, p = 0.0003) and 2-year OS was 54%, 47%, 24%, and 12% (log rank, p < 0.0001) for HR+/HER2+, HR-/HER2+, HR+/HER2-, and TN subtypes, respectively. On MVA, the TN subtype predicted for inferior DIC (HR 1.62, 95% CI 1.00-2.60, p = 0.049). The modified breast-Graded Prognostic Assessment (GPA) significantly predicted DIC and OS (both p < 0.001). CONCLUSIONS: Subtype is prognostic for OS and DIC for patients with BCBM treated with SRT.

Trastuzumab Emtansine (T-DM1) and stereotactic radiation in the management of HER2+ breast cancer brain metastases.

BACKGROUND: Due to recent concerns about the toxicity of trastuzumab emtansine (T-DM1) with stereotactic radiation, we assessed our institutional outcomes treating HER2-positive breast cancer brain metastases (BCBM) with T-DM1 and stereotactic radiation. METHODS: This is a single institution series of 16 patients with HER2-positive breast cancer who underwent 18 stereotactic sessions to 40 BCBM from 2013 to 2019 with T-DM1 delivered within 6 months. The Kaplan-Meier method was used to calculate overall survival (OS), local control (LC), distant intracranial control (DIC), and systemic progression-free survival (sPFS) from the date of SRS. A neuro-radiologist independently reviewed follow-up imaging. RESULTS: One patient had invasive lobular carcinoma, and 15 patients had invasive ductal carcinoma. All cases were HER2-positive, while 10 were hormone receptor (HR) positive. Twenty-four lesions were treated with stereotactic radiosurgery (SRS) to a median dose of 21 Gy (14-24 Gy). Sixteen lesions were treated with fractionated stereotactic radiation (FSRT) with a median dose of 25 Gy (20-30Gy) delivered in 3 to 5 fractions. Stereotactic radiation was delivered concurrently with T-DM1 in 19 lesions (48%). Median follow up time was 13.2 months from stereotactic radiation. The 1-year LC, DIC, sPFS, and OS were 75, 50, 30, and 67%, respectively. There was 1 case of leptomeningeal progression and 1 case (3%) of symptomatic radionecrosis. CONCLUSIONS: We demonstrate that stereotactic radiation and T-DM1 is well-tolerated and effective for patients with HER2-positive BCBM. An increased risk for symptomatic radiation necrosis was not noted in our series.

Hypofractionated stereotactic re-irradiation with pembrolizumab and bevacizumab in patients with recurrent high-grade gliomas: results from a phase I study.

BACKGROUND: Radiotherapy may synergize with programmed cell death 1 (PD1)/PD1 ligand (PD-L1) blockade. The purpose of this study was to determine the recommended phase II dose, safety/tolerability, and preliminary efficacy of combining pembrolizumab, an anti-PD1 monoclonal antibody, with hypofractionated stereotactic irradiation (HFSRT) and bevacizumab in patients with recurrent high-grade gliomas (HGGs). METHODS: Eligible subjects with recurrent glioblastoma or anaplastic astrocytoma were treated with pembrolizumab (100 or 200 mg based on dose level Q3W) concurrently with HFSRT (30 Gy in 5 fractions) and bevacizumab 10 mg/kg Q2W. RESULTS: Thirty-two patients were enrolled (bevacizumab-naïve, n = 24; bevacizumab-resistant, n = 8). The most common treatment-related adverse events (TRAEs) were proteinuria (40.6%), fatigue (25%), increased alanine aminotransferase (25%), and hypertension (25%). TRAEs leading to discontinuation occurred in 1 patient who experienced a grade 3 elevation of aspartate aminotransferase. In the bevacizumab-naïve cohort, 20 patients (83%) had a complete response or partial response. The median overall survival (OS) and progression-free survival (PFS) were 13.45 months (95% CI: 9.46-18.46) and 7.92 months (95% CI: 6.31-12.45), respectively. In the bevacizumab-resistant cohort, PR was achieved in 5 patients (62%). Median OS was 9.3 months (95% CI: 8.97-18.86) with a median PFS of 6.54 months (95% CI: 5.95-18.86). The majority of patients (n = 20/26; 77%) had tumor-cell/tumor-microenvironment PD-L1 expression <1%. CONCLUSIONS: The combination of HFSRT with pembrolizumab and bevacizumab in patients with recurrent HGG is generally safe and well tolerated. These findings merit further investigation of HFSRT with immunotherapy in HGGs.

Tumor Volume Dynamics as an Early Biomarker for Patient-Specific Evolution of Resistance and Progression in Recurrent High-Grade Glioma.

Recurrent high-grade glioma (HGG) remains incurable with inevitable evolution of resistance and high inter-patient heterogeneity in time to progression (TTP). Here, we evaluate if early tumor volume response dynamics can calibrate a mathematical model to predict patient-specific resistance to develop opportunities for treatment adaptation for patients with a high risk of progression. A total of 95 T1-weighted contrast-enhanced (T1post) MRIs from 14 patients treated in a phase I clinical trial with hypo-fractionated stereotactic radiation (HFSRT; 6 Gy × 5) plus pembrolizumab (100 or 200 mg, every 3 weeks) and bevacizumab (10 mg/kg, every 2 weeks; NCT02313272) were delineated to derive longitudinal tumor volumes. We developed, calibrated, and validated a mathematical model that simulates and forecasts tumor volume dynamics with rate of resistance evolution as the single patient-specific parameter. Model prediction performance is evaluated based on how early progression is predicted and the number of false-negative predictions. The model with one patient-specific parameter describing the rate of evolution of resistance to therapy fits untrained data ( R 2 = 0.70 ). In a leave-one-out study, for the nine patients that had T1post tumor volumes ≥1 cm3, the model was able to predict progression on average two imaging cycles early, with a median of 9.3 (range: 3-39.3) weeks early (median progression-free survival was 27.4 weeks). Our results demonstrate that early tumor volume dynamics measured on T1post MRI has the potential to predict progression following the protocol therapy in select patients with recurrent HGG. Future work will include testing on an independent patient dataset and evaluation of the developed framework on T2/FLAIR-derived data.

Current Treatment of Melanoma Brain Metastasis.

OPINION STATEMENT: With greater understanding of underlying biology and development of effective BRAF-targeted therapy and immunotherapy, along with remarkable advances in local treatment such as stereotactic radiosurgery, melanoma brain metastasis (MBM) is witnessing continually improving outcome, with 1-year overall survival rate approaching 85%. Given disease complexity and myriad treatment options, all patients with MBM should ideally be evaluated in a multidisciplinary setting to allow an individualized treatment approach based on prognostic groups, molecular classification, number and size of brain metastasis, and performance status. With improving outcome, pendulum has now swayed to focus more on effective treatment modalities with minimal neurological toxicity while maintaining quality of life. Surgery is usually considered in symptomatic and large MBMs, while stereotactic radiosurgery considered in 1-4 lesions, and now also being explored for up to 15 brain metastases for improved local control. The role of whole brain radiotherapy is diminishing given its neurocognitive toxicities and is reserved for patients with diffuse brain involvement. Cytotoxic chemotherapy has largely been ineffective without evidence for survival benefit. Immune checkpoint inhibitors have become the cornerstone of management for melanoma brain metastasis with durable intracranial tumor control and excellent toxicity profile. For patients with asymptomatic MBMs, ipilimumab and nivolumab have shown intracranial response near 60% and provides comparable clinical benefit in MBMs as for extracranial metastases. For patients with driver BRAF mutation, BRAFi-/MEKi-targeted agents are proven to be effective in MBM with high rate intracranial responses (44-59%). However, the durability of intracranial responses induced by BRAFi/MEKi seems to be shorter than that of extracranial disease. Emerging data support novel combination of systemic therapy and stereotactic radiosurgery, which appears to be safe and effective; however, potential benefits and risks should be evaluated prospectively. Promising ongoing trials will further expand therapeutic evidence in MBM, and patients should be encouraged to participate in clinical trials.

Outcomes and the Role of Primary Histology Following LINAC-based Stereotactic Radiation for Sarcoma Brain Metastases.

OBJECTIVES: The brain is a rare site for sarcoma metastases. Sarcoma's radioresistance also makes standard whole-brain radiotherapy less appealing. We hypothesize that stereotactic radiation techniques (stereotactic radiosurgery [SRS]/stereotactic fractionated radiotherapy [FSRT]) may provide effective local control. MATERIALS AND METHODS: This single-institution retrospective analysis evaluated our experience with linear acceleator-based SRS/FSRT for sarcoma brain metastases. Time to event analysis was estimated via Kaplan-Meier. Univariable/multivariable Cox regression analyses followed to assess the impact of patient and disease characteristics on outcomes. RESULTS: Between 2003 and 2018, 24 patients were treated with 34 courses of SRS/FSRT to 58 discrete lesions. The median age at first treatment was 57 years (range: 25 to 87 y). Majority of patients had concurrent lung metastases (n=21; 88%), diagnosed spindle cell sarcoma (n=15; 25%) or leiomyosarcoma (n=12; 21%) histology, and were treated with either SRS (n=43; median dose=19 Gy, range: 15 to 24 Gy) or FSRT (n=17; 3/5 fractions, median dose=25 Gy, range: 25 to 35 Gy). With a median follow-up after brain metastasis of 7.3 months, the 6 month/12 month local control, distant brain control, and overall survival of 89%/89%, 59%/34%, and 50%/38%, respectively. All local failures were of primary spindle cell histology (P<0.001), which was associated with poorer distant control (hazard ratio=25.8, 95% confidence interval: 3.1-536.4; P=0.003) on univariable analysis, and OS (hazard ratio=7.1, 95% confidence interval: 2.0-26.1; P=0.003) on multivariable analysis. CONCLUSIONS: This is the largest patient cohort with sarcoma brain metastases treated with SRS/FSRT, it provides durable local control, despite a reputation for radioresistance. Further prospective evidence is required to determine the impact of primary histology on control and survival following brain metastasis diagnosis.

Management of brain metastases in breast cancer: a review of current practices and emerging treatments.

PURPOSE: Breast cancer brain metastases (BCBM) are becoming an increasingly common diagnosis due to improved systemic control and more routine surveillance imaging. Treatment continues to require a multidisciplinary approach managing systemic and intracranial disease burden. Although, improvements have been made in the diagnosis and management of BCBM, brain metastasis patients continue to pose a challenge for practitioners. METHODS: In this review, a group of medical oncologists, radiation oncologists, radiologists, breast surgeons, and neurosurgeons specializing in the treatment of breast cancer reviewed the available published literature and compiled a comprehensive review on the current state of BCBM. RESULTS: We discuss the pathogenesis, epidemiology, diagnosis, treatment options (including systemic, surgical, and radiotherapy treatment modalities), and treatment response evaluation for BCBM. Furthermore, we discuss the ongoing prospective trials enrolling BCBM patients and their biologic rationale. CONCLUSIONS: BCBM management is an increasing clinical concern. Multidisciplinary management combining the strengths of surgical, systemic, and radiation treatment modalities with prospective trials incorporating knowledge from the basic and translational sciences will ultimately lead to improved clinical outcomes for BCBM patients.