Faculty Bibliography

BACKGROUND AND OBJECTIVES/OBJECTIVE:Bone metastases, affecting more than 4.8% of patients with cancer annually, and particularly spinal metastases require urgent intervention to prevent neurological complications. However, the current process of manually reviewing radiological reports leads to potential delays in specialist referrals. We hypothesized that natural language processing (NLP) review of routine radiology reports could automate the referral process for timely multidisciplinary care of spinal metastases. METHODS:We assessed 3 NLP models-a rule-based regular expression (RegEx) model, GPT-4, and a specialized Bidirectional Encoder Representations from Transformers (BERT) model (NYUTron)-for automated detection and referral of bone metastases. Study inclusion criteria targeted patients with active cancer diagnoses who underwent advanced imaging (computed tomography, MRI, or positron emission tomography) without previous specialist referral. We defined 2 separate tasks: task of identifying clinically significant bone metastatic terms (lexical detection), and identifying cases needing a specialist follow-up (clinical referral). Models were developed using 3754 hand-labeled advanced imaging studies in 2 phases: phase 1 focused on spine metastases, and phase 2 generalized to bone metastases. Standard McRae's line performance metrics were evaluated and compared across all stages and tasks. RESULTS:In the lexical detection, a simple RegEx achieved the highest performance (sensitivity 98.4%, specificity 97.6%, F1 = 0.965), followed by NYUTron (sensitivity 96.8%, specificity 89.9%, and F1 = 0.787). For the clinical referral task, RegEx also demonstrated superior performance (sensitivity 92.3%, specificity 87.5%, and F1 = 0.936), followed by a fine-tuned NYUTron model (sensitivity 90.0%, specificity 66.7%, and F1 = 0.750). CONCLUSION/CONCLUSIONS:An NLP-based automated referral system can accurately identify patients with bone metastases requiring specialist evaluation. A simple RegEx model excels in syntax-based identification and expert-informed rule generation for efficient referral patient recommendation in comparison with advanced NLP models. This system could significantly reduce missed follow-ups and enhance timely intervention for patients with bone metastases.

BACKGROUND AND OBJECTIVES/OBJECTIVE:The therapeutic approach for small- and medium-sized vestibular schwannoma (VS) with normal hearing function remains controversial, with limited comparative data regarding hearing outcomes after stereotactic radiosurgery (SRS) or observation (OBS). We evaluated the serviceable hearing preservation, loss of American Academy of Otolaryngology-Head and Neck Surgery class A hearing, and tumor control (TC) across individuals with Koos grade I and II VSs and normal hearing at presentation who underwent SRS or OBS. METHODS:In this multicenter international study, we retrospectively analyzed the hearing, radiological, and neurological outcomes of patients who underwent SRS (SRS group) or OBS (OBS group). The cohorts were matched using propensity scores based on age, sex, tumor volume, pure-tone average, and speech discrimination score at a 1:1 ratio without replacement. RESULTS:After matching, each group comprised 57 patients. The median follow-up was 49 and 37 months for the SRS and the OBS groups, respectively (P = .3). The 5- and 9-year serviceable hearing preservation rates in the SRS group were 76.2% and 42.4% vs 56.1% and 16.8% in the OBS group (P = .17). Class A preservation occurred in 57.9% (33/57) of the SRS and 52.6% (30/57) of the OBS cohorts (P = .70). Regarding the TC rates, SRS was associated with significantly higher TC rates (P < .0001). CONCLUSION/CONCLUSIONS:We found that SRS is significantly superior regarding TC and provided noninferior hearing outcomes compared with OBS in VS patients with American Academy of Otolaryngology-Head and Neck Surgery class A hearing at presentation. Therefore, we suggest performing SRS in individuals with VS and normal hearing function.

BACKGROUND AND OBJECTIVES/OBJECTIVE:Since the introduction of the Leksell Gamma Knife to North America in 1987, stereotactic radiosurgery (SRS) has increasingly been used for patients with intracranial meningiomas. We evaluated the evolving application and outcomes of meningioma patients managed with both primary and adjuvant SRS during a 35-year interval. METHODS:The authors reviewed the outcomes of meningioma patients (1229 female, 69.8%; 2220 tumors) who underwent single-fraction SRS from August 1987 to March 2022 and who had a minimum of 6-month follow-up. The rates of treated tumor control and overall survival up to 20 years after SRS were measured. Risk factors analyzed included age, sex, tumor volume, margin dose, Ki-67, anatomical location, and pre-SRS surgical resection. RESULTS:Primary SRS showed superior tumor control compared with adjuvant SRS after previous resection. Overall, 191 of 2220 patients (8.6%) had local progression at last follow-up with the 5-year, 10-year, 15-year, and 20-year tumor control rates were 92.1%, 88.3%, 84.1%, and 81.1%, respectively. The median overall survival after SRS was 17.4 years, and 2.6% of patients died related to meningioma progression. Patients treated so that ≥60% of the tumor received at least 16 Gy demonstrated significantly superior tumor control. Fifty-eight patients (3.3%) experienced symptomatic adverse radiation effects after SRS. CONCLUSION/CONCLUSIONS:SRS provided excellent local tumor control rates that extended beyond 20 years. Primary SRS was an effective strategy for patients with unresected or known WHO grade I meningiomas. Adjuvant SRS was an important option to enhance tumor control and survival in patients with residual or progressive tumors after resection.

OBJECTIVE:To characterize patient outcomes after primary stereotactic radiosurgery (SRS) for the management of sporadic facial nerve schwannoma. STUDY DESIGN/METHODS:Retrospective cohort study. SETTING/METHODS:Six tertiary referral centers across the United States and United Kingdom. PATIENTS/METHODS:Adults undergoing SRS from 2000 through 2023 for sporadic facial nerve schwannoma along any segment of the facial nerve were included. Patients with NF2-related schwannomatosis were excluded. INTERVENTION/METHODS:Stereotactic radiosurgery. MAIN OUTCOME MEASURE/METHODS:Long-term tumor control. RESULTS:Among 60 patients meeting inclusion, the median age at SRS was 52 years (IQR: 41 to 64) with a median tumor size of 19.5 mm (IQR: 14.7 to 22.8). Tumors commonly involved the internal auditory canal (73%), cisternal (49%), geniculate/labyrinthine (47%), and tympanic segments (22%). Two patients experienced SRS failure and underwent salvage treatment; salvage-free survival rates (95% CI; number still at risk) at 1, 3, 5, and 10 years after SRS were 100% (100 to 100; 55), 100% (100 to 100; 36), 100% (100 to 100; 18), and 87% (72 to 100; 9), respectively. Among 31 (52%) patients with House-Brackmann (HB) grade I facial function at presentation, only 6 demonstrated worse facial function at a median of 3.2 years (IQR: 1.7 to 6.6) after SRS. Of 18 patients with serviceable hearing (AAO-HNS class A/B) at SRS, 13 maintained serviceable hearing at a median of 1.0 years (IQR: 0.5 to 4.9) of post-SRS audiometric follow-up. CONCLUSIONS:Durable tumor control after primary SRS for sporadic facial nerve schwannoma is achieved in most patients. Among those with HB grade I facial function at presentation, treatment with SRS harbors limited additional risk of facial paresis beyond observation alone.

BACKGROUND:Intracranial solitary fibrous tumors (SFTs) are rare, aggressive neoplasms with high local recurrence. This study evaluates the efficacy and prognostic factors of single-fraction stereotactic radiosurgery (SRS). METHODS:This multicenter retrospective study included 107 patients (253 SFTs) treated with single-fraction SRS at 18 centers (1989-2024). We analyzed local control (LC), intracranial tumor control (ITC), overall tumor control (OTC), progression-free survival (PFS), disease-specific survival (DSS), and overall survival (OS). Cox regression identified prognostic factors. RESULTS:Median follow-up was 33 months. LC was 68.4% (5-yr: 56.8% and 10-yr: 38.8%). ITC 54.2% (5-yr: 48.5%) and OTC 50.5% (5-yr: 44.0%). PFS was 56.3% and 30.2% at 5 and 10 years, respectively. DSS remained high at 89.7% (5-yr) and 79.7% (10-yr), while OS was 79.3% (5-yr) and 55.2% (10-yr). Independent predictors of LC included recurrent vs. metastatic SFTs (HR: 1.96, p = 0.028), margin dose ≤16 Gy (HR: 2.35, p = 0.006), larger tumor volume (HR: 1.05, p < 0.001), and longer diagnosis-to-SRS duration (HR : 1.02, p < 0.001). Older age (HR: 1.03, p = 0.014) and longer resection-to-SRS duration (HR: 1.02, p = 0.018) predicted worse ITC. Age significantly affected OS (HR: 1.06, p < 0.001) and PFS (HR: 1.03, p = 0.037). Longer diagnosis-to-SRS (HR: 1.03, p = 0.002) and resection-to-SRS durations (HR : 1.02, p = 0.030) predicted worse PFS. KPS score >70 predicted better outcomes across ITC, OTC, DSS and OS. Radiation-related adverse effects occured in 2.8%. CONCLUSION/CONCLUSIONS:SRS offers reasonable tumor control and favorable long-term survival in the adjuvant and salvage setting for patients with residual, recurrent, or metastatic intracranial SFTs. Key prognostic factors included tumor volume, recurrence status, and timing of SRS.

BACKGROUND AND OBJECTIVES/OBJECTIVE:Stereotactic radiosurgery (SRS) has been increasingly employed in the multimodal management of primary central nervous system lymphoma. Here, we evaluate the outcomes of SRS for the treatment of primary central nervous system lymphoma through a multicenter, international cohort study. METHODS:A multicenter, retrospective cohort study was conducted through the International Radiosurgery Research Foundation. Subgroups were defined according to treatment setting: up-front (primary treatment), boost (SRS after consolidative chemotherapy with or without whole-brain radiotherapy), and relapsed/refractory (recurrent/progressive disease after first-line therapy). The primary end point was local tumor control. Time-to-event analysis was conducted using the Kaplan-Meier method. Variables associated with local control were assessed using the Cox proportional hazard modeling. RESULTS:Fifty-four patients with 127 tumor sites were included in this analysis. Actuarial 12-month local and distant control rates for the entire cohort were 75.7% and 63.7%, respectively, with a median overall survival (OS) of 18 months (range: 1-176). Actuarial 12-month local control rates were significantly different at 95.6%, 78.3%, and 42.1% (P < .0001) for the up-front, relapsed/refractory groups, and boost cohorts, respectively. OS across all cohorts were similar with 12-month OS rates of 55.2%, 51.3%, and 53.5% for the up-front, relapsed/refractory, and boost cohorts, respectively. Rates of radiation necrosis were 18.5%, 20.8%, 15.4%, and 11.8% for the entire cohort, relapsed/refractory, boost, and up-front cohorts, respectively. Diminished OS was significantly associated with treatment volumes >27 cm3 (hazard ratio: 2.5, P = .04). CONCLUSION/CONCLUSIONS:SRS shows promising local tumor control rates for up-front and relapsed cohorts. Despite this, distant tumor progression limits total tumor control and may adversely affect OS.

BACKGROUND AND OBJECTIVES/OBJECTIVE:Gynecological cancers represent 10% to 15% of cancers in women, but brain metastases (BM) are uncommon, with limited evidence regarding their management. This study investigates the role of stereotactic radiosurgery (SRS) for BM from primary gynecological cancers. METHODS:Institutions of the International Radiosurgery Research Foundation participated in this study. Inclusion criteria required histological diagnosis of epithelial ovarian, cervical, or endometrial cancer, SRS between 2000 and 2020, and at least 1 imaging or clinical follow-up. RESULTS:A total of 276 patients having SRS for 977 BM were included. Median age at SRS was 62 years (IQR, 55-70). Primary cancer origin was ovarian in 128 (46%), cervical in 43 (16%), and endometrial in 105 patients (38%). Median Karnofsky Performance Scale was 80%, and systemic disease was active in 124 (45%) of patients. A median of 1 metastasis was treated (IQR, 1-3) per patient. Median individual metastasis volume was 0.27 cc (IQR, 0.05-1.59 cc). The majority (91%) received single-fraction SRS, using a median margin dose of 18 Gy (IQR, 16-20 Gy). Actuarial overall survival was 77%, 65%, and 44% at 6, 12, and 24 months, respectively. Predictors of worsened survival included older age, cervical and endometrial primary, previous whole-brain radiation therapy (WBRT), active systemic disease, worsened Karnofsky Performance Scale, absence of subsequent surgery, and increasing number of BM. Actuarial local control was 94% at 6 months, 89% at 12 months, and 78% at 24 months. Previous SRS or WBRT, tumor bed treatment, and cervical histology increased the risk of local failure. New remote BM and leptomeningeal dissemination occurred in 44% and 11% of patients, respectively. Adverse radiation effects (ARE) occurred in 13% of cases but were symptomatic in only 3%. Previous WBRT or SRS and increased tumor diameter increased the risk of ARE. CONCLUSION/CONCLUSIONS:SRS is an effective management for BM from gynecological cancers with low risks of symptomatic ARE.

BACKGROUND AND OBJECTIVES/OBJECTIVE:Single-session stereotactic radiosurgery (SRS) has limited role for large arteriovenous malformations (AVM). Volume-staged SRS (VS-SRS) is used to optimize outcomes, but studies reporting results are limited. METHODS:This multicenter retrospective cohort of 378 patients from 21 centers reports results of VS-SRS for the entire AVM nidus. We report favorable outcome, obliteration, hemorrhage, and permanent symptomatic adverse radiation effect rates. RESULTS:The median age was 31 years (IQR: 19-44) at the first volume stage, with patients treated in 2-4 stages. The median total nidus volume was 21 cm3 (IQR: 13.9-30.1 cm3), and a median prescription dose of 17 Gy (IQR: 16-18 Gy) was used. The median radiographic and clinical follow-up were 48 and 55 months, respectively. Seventy-seven patients (20.4%) had a favorable outcome, with the 3-year and 5-year rates being 3.9% and 18%, respectively. 127 patients (33.6%) achieved obliteration, with the 3-year and 5-year rates being 6.8% and 26%, respectively. Obliteration rates of AVMs <15 cm3 were 81% and 31%, respectively. The latency period hemorrhage incidence rate was 3.02 cases per 100 patient-years; 52 patients (13.8%) had a bleed. Seventy-two patients (19%) had symptomatic adverse radiation effect; in 38 patients (10.1%), these were permanent. Total nidus volume, prescription dose at first stage, diffuse nidus, and prior hemorrhage were all independent affecting outcome rates. CONCLUSION/CONCLUSIONS:VS-SRS can be used to treat large AVMs as a standalone treatment. Obliteration rates and favorable outcomes are lower than that with smaller AVMs, and repeat treatment is often required. Optimizing treatment plans, by increasing prescription doses, reducing treatment volume at each stage, and increasing the number of stages, may lead to better outcomes.

BACKGROUND AND OBJECTIVES/OBJECTIVE:The Congress of Neurological Surgeons Self-Assessment for Neurological Surgeons questions are widely used by neurosurgical residents to prepare for written board examinations. Recently, these questions have also served as benchmarks for evaluating large language models' (LLMs) neurosurgical knowledge. LLMs show significant promise for transforming neurosurgical practice; however, they are susceptible to in-text distractions and confounding factors. Given the increasing use of generative artificial intelligence and ambient dictation technologies, clinical text is at a larger risk for the inclusion of extraneous details. The aim of this study was to assess the performance of state-of-the-art LLMs on neurosurgery board-like questions and to evaluate their robustness to the inclusion of distractor statements. METHODS:A comprehensive evaluation was conducted using 28 state-of-the-art LLMs. These models were tested on 2904 neurosurgery board examination questions derived from the Congress of Neurological Surgeons Self-Assessment for Neurological Surgeons. In addition, the study introduced a distraction framework to assess the fragility of these models. The framework incorporated simple, irrelevant distractor statements containing polysemous words with clinical meanings used in nonclinical contexts to determine the extent to which such distractions degrade model performance on standard medical benchmarks. RESULTS:Six of the 28 tested LLMs achieved board-passing outcomes, with the top-performing models scoring over 15.7% above the passing threshold. When exposed to distractions, accuracy across various model architectures was significantly reduced-by as much as 20.4%-with 1 model failing that had previously passed. Both general-purpose and medical open-source models experienced greater performance declines compared with proprietary variants when subjected to the added distractors. CONCLUSION/CONCLUSIONS:While current LLMs demonstrate an impressive ability to answer neurosurgery board-like examination questions, their performance is markedly vulnerable to extraneous, distracting information. These findings underscore the critical need for developing novel mitigation strategies aimed at bolstering LLM resilience against in-text distractions, particularly for safe and effective clinical deployment.

BACKGROUND AND OBJECTIVES/OBJECTIVE:Repeat stereotactic radiosurgery (SRS) is a noninvasive option for recurrent vestibular schwannoma (VS). This study evaluates outcomes in patients with long-term follow-up. METHODS:This retrospective multicenter study analyzed 81 patients with recurrent unilateral sporadic VS after initial SRS, with ≥12 months of follow-up. Outcomes included tumor control, hearing preservation, cranial nerve function, and adverse radiation effects (ARE). Kaplan-Meier and Cox regression identified factors affecting outcomes. RESULTS:The median age at the second SRS was 60 years, with a median interval of 58 months between procedures. The median margin doses were 12.0 Gy (single-fraction), 17.25 Gy (3-fraction), and 25 Gy (5-fraction). Tumor control was achieved in 69 patients (85.2%), with 5- and 10-year local control rates of 82% and 76.5%, respectively. Significant predictors of local failure included tumor volume >2.2 cm3 (area under the curve = 0.757, P = .018), prescription biological effective dose (BED) ≤70.3 Gy (hazard ratio [HR]: 0.89, P = .003), and interval between treatments >27.5 months (HR: 1.02, P = .015). In single-fraction SRS, higher prescription dose reduced failure risk (HR: 0.31, P = .002) with a margin dose ≥12 Gy being critical for improved tumor control (P < .001). Serviceable hearing was retained in 12 of 18 cases (66.7%), and facial nerve function was preserved in 72 of 80 cases (90%). ARE occurred in 11 patients (13.6%), most commonly perilesional edema (63.7%). ARE correlated with higher brainstem maximum BED in the entire cohort (HR: 1.02, P = .016) and in single-fraction SRS (HR: 1.02, P = .006). Pseudoprogression (9.8%) was linked to younger age (HR: 0.88, P = .023) and shorter time between SRS (HR: 0.87, P = .012). CONCLUSION/CONCLUSIONS:Repeat SRS is an effective option for recurrent sporadic VS, offering high tumor control and functional preservation. Outcomes depend on age, interval between treatments, tumor volume, and BED. With careful planning, adverse effects are rare and typically transient.