Neuro-oncology最新文献

{"title":"EGFRvIII: An old oncogenic target gets a model built for the immunotherapy era.","authors":"Sarah A Best, James R Whittle, David D Eisenstat","doi":"10.1093/neuonc/noag249","DOIUrl":"https://doi.org/10.1093/neuonc/noag249","url":null,"abstract":"","PeriodicalId":19377,"journal":{"name":"Neuro-oncology","volume":" ","pages":""},"PeriodicalIF":13.1,"publicationDate":"2026-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149137223","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatially restricted iron-metabolism-associated stress programs prime invasive mesenchymal glioblastoma states in human tissue-context models.","authors":"Christophe Petry, Ambra-Stella Boecke, Aso Omer Mohammed, Nishita S Ghariwala, Vatsal D Jariwala, Mohammad Al Shhab, Oliver Buchholz, Ioannis Vasilikos, Julia M Nakagawa, Andreas Vlachos, Jürgen Grauvogel, Mukesch J Shah, Roland Rölz, Marco Prinz, Ulrich G Hofmann, Jürgen Beck, Kevin Joseph, Vidhya Madapusi Ravi","doi":"10.1093/neuonc/noag220","DOIUrl":"https://doi.org/10.1093/neuonc/noag220","url":null,"abstract":"<p><strong>Background: </strong>Glioblastoma (GBM) recurrence is driven by invasive tumor cells that infiltrate surrounding brain tissue and evade surgical and therapeutic eradication. While dysregulated iron handling is a well-recognized feature of malignant cancers, its contribution to invasive GBM cell states in the physiological context of human cortical tissue remains poorly defined.</p><p><strong>Methods: </strong>We performed multiregional single-nucleus RNA sequencing of human GBM specimens encompassing tumor core, tumor periphery, and infiltrated cortex to map iron-associated transcriptional programs validated by spatial transcriptomics. Correlative analyses were combined with functional validation using patient-derived GBM cell lines and human organotypic cortical slice cultures exposed to noncytotoxic iron supplementation to model iron-rich tumor microenvironments.</p><p><strong>Results: </strong>Malignant cells from the tumor core exhibited coordinated upregulation of iron uptake and storage pathways alongside invasion-associated gene programs. At single-cell resolution, iron metabolism and invasion signatures were correlated, defining a core-enriched malignant subpopulation with mesenchymal-like transcriptional identity, stress-adaptive features, and angiogenic signaling that aggregate in specific spatial niches. Functionally, iron exposure altered migration in a cell-state-dependent manner and increased tumor growth and invasion in human cortical slice cultures. Increased VIM, MMP9, and HIF1A expression was consistent with activation of mesenchymal-like and stress-response programs, although mechanistic dependence was not tested.</p><p><strong>Conclusion: </strong>Iron-handling and invasion-associated programs co-occur in a core-enriched mesenchymal-like GBM state. In patient-derived cell and human cortical slice models, noncytotoxic particulate and soluble iron altered migration in a cell-state-dependent manner and increased tissue invasion. These findings support iron availability as a component of the microenvironment associated with invasive GBM phenotypes, while its necessity and underlying mechanism remain to be established.</p>","PeriodicalId":19377,"journal":{"name":"Neuro-oncology","volume":" ","pages":""},"PeriodicalIF":13.1,"publicationDate":"2026-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149128262","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Treatment deintensification strategies in primary central nervous system germinoma.","authors":"Matthew D Hall, Mohamed S Abdelbaki","doi":"10.1093/neuonc/noag244","DOIUrl":"https://doi.org/10.1093/neuonc/noag244","url":null,"abstract":"","PeriodicalId":19377,"journal":{"name":"Neuro-oncology","volume":" ","pages":""},"PeriodicalIF":13.1,"publicationDate":"2026-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149128299","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An evidence-based workflow for integrating molecular testing with histopathology in the workup of meningiomas.","authors":"Harrshavasan Congivaram, Vineeth Thirunavu, Lucas Santana-Santos, Pouya Jamshidi, Jared T Arendhsen, Rudolph J Castellani, Daniel J Brat, Lee A D Cooper, Marina A Ayad, William C Chen, Khizar R Nandoliya, Rahul K Chaliparambil, Rishi Jain, Mateo Gomez, Drew Duckett, Madina Sukhanova, Heather Smith, Matthew McCord, Alicia Steffens, Jordain Walshon, Kathleen McCortney, Wenxia Wang, Xinyan Lu, Amy B Heimberger, David R Raleigh, Stephen T Magill, Mark W Youngblood, Craig M Horbinski","doi":"10.1093/neuonc/noag217","DOIUrl":"https://doi.org/10.1093/neuonc/noag217","url":null,"abstract":"<p><strong>Background: </strong>Meningiomas are graded using histologic and molecular criteria according to the World Health Organization (WHO) classification. However, there is scant evidence considering the specific value of each individual atypical histological feature that contributes to WHO grading. There is also little evidence-based guidance as to which cases should receive molecular testing.</p><p><strong>Methods: </strong>Detailed histological information was collected for 1058 resected meningiomas, including cases with genomic DNA methylation and copy number variant (CNV) data, next-generation sequencing, and RNA risk score. Integrated analyses were conducted using known prognostic variables, including WHO Grade, Ki-67, mitotic index (MI), and atypical histologic features (necrosis, hypercellularity, macronucleoli, sheeting, and small cells).</p><p><strong>Results: </strong>Necrosis, elevated MI, and increased WHO grade correlated with specific DNA methylation patterns, RNA risk groups, and chromosomal CNVs. Predictive analysis identified MI as the strongest single correlate, and suggested the presence of ≥ 2 atypical features as an optimal threshold for recurrence prediction. Sheeting architecture was the only independent histologic correlate of recurrence. Integration of histologic and molecular features suggests that tumors with high MI have poor prognosis regardless of molecular profile. Further analysis of low MI tumors revealed specific predictors of unfavorable molecular profiles.</p><p><strong>Conclusions: </strong>These data indicate that some atypical histopathologic features are more powerful than others, and can be used on initial screening to increase the yield of molecular profiling of meningiomas. Furthermore, we propose that molecular testing may actually have greatest added value in cases with low, not high, MI.</p>","PeriodicalId":19377,"journal":{"name":"Neuro-oncology","volume":" ","pages":""},"PeriodicalIF":13.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149112449","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Nuclear export as a therapeutic vulnerability in ZFTA-RELA ependymoma.","authors":"Siri M Ippagunta, Amir Arabzade, Srinidhi Varadharajan, Erik Emanus, Amanda Bland, Alisha Kardian, Nicholas Laboe, Hua Sun, Tuyu Zheng, Amelia Hancock, Mark Zhai, Emma Caesar, Maxwell Park, Wilda Orisme, Melissa Johnson, Brittany Rodgers, Gregory H Turner, Jeffrey Steinberg, Burgess Freeman, Martine F Roussel, Anang Shelat, Richard W Kriwacki, Thomas E Merchant, Benjamin Deneen, Michael D Taylor, Sameer Agnihotri, Shondra Pruett-Miller, Jacob A Steele, Anoop Murthy Kavirayani, David W Ellison, Arzu Onar-Thomas, Giles W Robinson, Stephen C Mack, Kelsey C Bertrand","doi":"10.1093/neuonc/noag100","DOIUrl":"10.1093/neuonc/noag100","url":null,"abstract":"<p><strong>Background: </strong>Ependymoma (EPN) is the third most common pediatric brain tumor with no targeted therapies available to patients. In supratentorial ependymoma, the most frequent driver alteration is a gene fusion between ZFTA and RELA (denoted ZR), which leads to constitutive localization of ZR in the nucleus. Because ZR is not currently druggable, we tested whether ZR expression leads to aberrant protein interactions that could represent therapeutic vulnerabilities.</p><p><strong>Methods: </strong>Using CRISPR-Cas9 pooled screening, we identified many novel druggable ZR-interacting proteins including XPO1, CARM1, SMARCA4, and CDK1. We focused on the nuclear export protein (XPO1), given the ability of most XPO1 inhibitors (ie, selinexor) to cross the blood-brain barrier, FDA approval, and have documented safety profiles in children.</p><p><strong>Results: </strong>We found that specific nuclear ZR levels are needed for cell proliferation and are regulated by XPO1. Increased ZR accumulation in the nucleus does not increase oncogenic gene expression but drives tumor cells out of the cell cycle, as compared to a defective ZR DNA-binding mutant. Treatment of ZR-driven patient-derived mouse models with selinexor impairs cell growth and extends survival of animals in vivo. The combination of selinexor treatment with gemcitabine and ribociclib (used in a clinical trial for relapsed EPN at St. Jude Children's Research Hospital [SJDAWN]) further extends mouse survival.</p><p><strong>Conclusion: </strong>Our findings demonstrate that ZR-interacting proteins constitute therapeutic leads and that XPO1 is critical for titrating \"goldilocks\" levels of ZR nuclear expression. We identify a novel combination therapy of selinexor, gemcitabine, and ribociclib that may be immediately translated into clinical trials for EPN patients that are currently without targeted treatments.</p>","PeriodicalId":19377,"journal":{"name":"Neuro-oncology","volume":" ","pages":"2472-2484"},"PeriodicalIF":13.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13498431/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147818096","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The clinical and molecular landscape of thalamic glioma.","authors":"Linpeng Zhang, Hesong Wang, Guangrui Huang, Xueling Qi, Angel M Carcaboso, Chen Wang, Jing Feng, Song Han, Zhilong Cheng, Zhong Ma, Yakun Yang, Heng Xu, Minshuo Chen, Yongli Li, Xufei Zhang, Gang Cui, Jiacheng Lyu, Zejun Duan, Anlong Xu, Changxiang Yan","doi":"10.1093/neuonc/noag146","DOIUrl":"10.1093/neuonc/noag146","url":null,"abstract":"<p><strong>Background: </strong>Thalamic gliomas (TGs) remain a formidable clinical challenge for accurate diagnosis and effective therapy. This study aims to refine molecular diagnosis and surgical management of TGs by integrating clinicopathological and multiomics data.</p><p><strong>Methods: </strong>We analyzed 106 TGs, comprising 65 diffuse midline glioma (DMG) and 41 non-DMG, using genomic profiling, single-cell RNA sequencing, and orthogonal assays. Surgical outcomes were evaluated using rigorous causal inference frameworks, including propensity score matching and difference-in-differences analyses.</p><p><strong>Results: </strong>Genomic profiling revealed striking mutual exclusivity between H3F3A mutations and CDKN2A/B (9p21) loss. Single-cell and orthogonal assays validated that 9p21 loss emerged as a highly specific adjunctive marker favoring non-DMG over DMG. Causal inference analyses consistently demonstrated that microsurgical third ventriculostomy (mTV) confers profound protection against postoperative hydrocephalus (risk reduction >60%). Multivariate analysis identified the transfrontal approach as an independent predictor of postoperative motor deficits. Survival analysis showed that the median overall survival of this cohort following surgical resection was 21.0 months. Extent of resection and tumor mutational burden emerged as the principal independent determinants of prognosis, whereas postoperative Karnofsky performance status showed survival stratification in univariable/subgroup analyses but was not independently significant in the multivariable model.</p><p><strong>Conclusions: </strong>In this 106-patient cohort study, we delineated the somatic mutational landscape of TG and showed that 9p21 loss is mutually exclusive with H3F3A mutations, providing a highly specific adjunctive marker for TG. We further demonstrated that maximal surgical resection confers a survival benefit and intraoperative mTV serves as a standardized treatment workflow to optimize care for TG.</p>","PeriodicalId":19377,"journal":{"name":"Neuro-oncology","volume":" ","pages":"2417-2434"},"PeriodicalIF":13.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13635587/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148352706","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Beyond dose escalation in glioblastoma: Lessons learned from INTRAGO-II and the case for targeting tumor biology.","authors":"Julian Philipp Layer, Matthias Schneider, Michael Hölzel, Ulrich Herrlinger, Frank Winkler","doi":"10.1093/neuonc/noag166","DOIUrl":"10.1093/neuonc/noag166","url":null,"abstract":"","PeriodicalId":19377,"journal":{"name":"Neuro-oncology","volume":" ","pages":"2294-2295"},"PeriodicalIF":13.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13635592/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148990909","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Gadolinium-free MR perfusion imaging based on generative adversarial network for primary intracranial tumor diagnosis: A multi-center study.","authors":"Guoqi Lin, Wangbin Ding, Yihua Chen, Lukui Xiong, Junhuan Hong, Pingping He, Yuwei Pan, Minxuan Tian, Jiahao Tao, Wei Guo, Yan Su, Ming Chen, Dejun She","doi":"10.1093/neuonc/noag144","DOIUrl":"10.1093/neuonc/noag144","url":null,"abstract":"<p><strong>Background: </strong>Accurate evaluation of tumor vascularity is essential in neuro-oncology, but conventional cerebral blood volume (CBV) mapping depends on gadolinium-based contrast agents. The aim was to develop a gadolinium-free model to synthesize CBV (CBVsyn) maps from non-contrast MRI (NC-MRI).</p><p><strong>Methods: </strong>This multicenter retrospective study analyzed 1227 MRI examinations in patients with primary intracranial tumors, comprising an internal cohort (train/validation/test 438/126/63) and 2 external cohorts (UPENN-GBM Cohort 464; UCSD-PTGBM cohort 136). A tumor-aware generative adversarial network (TA-GAN) was implemented to synthesize CBVsyn maps from various NC-MRI combinations. Image quality was evaluated using structural similarity index (SSIM), peak signal-to-noise ratio (PSNR), and a 4-point Likert scale. The correlation between pathologic markers and CBV parameters was conducted. Diagnostic performance was compared between CBVsyn and DSC-PWI-derived CBV (CBVreal) maps. The incremental diagnostic value of CBVsyn was evaluated.</p><p><strong>Results: </strong>TA-GAN utilizing T2-weighted and diffusion-weighted imaging achieved best performance in CBVsyn synthesis (PSNRGlobal=25.52 ± 1.33dB, SSIMGlobal=67.53 ± 5.88; both P < .05), with 95.34% maps being diagnostically acceptable. CBVsyn parameters showed positive correlation with CD105 markers (r = 0.49, P = .03). CBVsyn maps demonstrated comparable AUC with CBVreal maps for intracranial tumors grading (0.61 vs. 0.66, P = .51), pediatric glioma (0.60 vs. 0.57; P = .57), adult glioma (0.68 vs. 0.73; P = .56), and IDH1 genotype (0.66 vs. 0.81; P = .03). NC-MRI with CBVsyn assistance improved diagnostic accuracy (84.48% vs.75.86%) and consistency (κ  = 0.83 vs. 0.62) for glioma grading compared with NC-MRI alone.</p><p><strong>Conclusion: </strong>As a gadolinium-free alternative, CBVsyn maps synthesized by TA-GAN correlate with pathologic findings and demonstrate promise for characterizing primary intracranial tumors.</p>","PeriodicalId":19377,"journal":{"name":"Neuro-oncology","volume":" ","pages":"2500-2513"},"PeriodicalIF":13.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13635576/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148331693","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A MYCN-GAL-SREBP1 lipogenic axis drives proliferation in silent corticotroph adenomas.","authors":"Changxi Han, Junchen Wu, Jing Xie, Jun Li, Zongze He, Bo Zhang, Yihan Dong, Qiuyue Fang, Jingwei Ye, Zhuoyi Tang, Can Xu, Jin Yang, Yangyang Wang, Mingjun Gao, Jin Zhang, Huan Xiong, Zhe Bao Wu, Ruxiang Xu, Shaojian Lin","doi":"10.1093/neuonc/noag141","DOIUrl":"10.1093/neuonc/noag141","url":null,"abstract":"<p><strong>Background: </strong>Silent corticotroph adenomas (SCAs) are an aggressive pituitary neuroendocrine tumor (PitNET) subtype lacking effective medical therapies and showing a high rate of recurrence. The molecular mechanisms driving their proliferation remain poorly understood. Although metabolic reprogramming is a hallmark of cancer, neither lipid metabolism nor its regulatory pathways have been systematically investigated in SCAs.</p><p><strong>Methods: </strong>We performed integrative analyses of bulk and single-cell RNA sequencing datasets from SCAs, functioning corticotroph adenomas (FCAs), and normal pituitary tissue to characterize GAL expression and associated signaling pathways. Mechanistic studies employed GAL gain- and loss-of-function models, RNA sequencing, luciferase reporter assays, lipidomics, and pharmacological inhibition. Regulation of GAL by MYCN was assessed through promoter transactivation assays. The therapeutic efficacy of SREBP1 inhibition (fatostatin) was evaluated both in vitro and in vivo.</p><p><strong>Results: </strong>We identified a marked downregulation of galanin (GAL) in SCAs and demonstrated that GAL loss promotes tumor cell proliferation. GAL deficiency activated the PI3K-Akt-mTOR signaling cascade, resulting in increased activity of SREBP1, a key transcriptional regulator of lipogenesis. Enhanced fatty-acid synthesis provided metabolic support for SCA growth. We further uncovered a linear regulatory axis in which MYCN directly upregulates GAL; however, MYCN downregulation in SCAs suppresses GAL expression, thereby enabling SREBP1-driven lipogenesis. Pharmacological inhibition of SREBP1 with fatostatin reduced lipogenesis and significantly suppressed SCA growth in vitro and in vivo.</p><p><strong>Conclusions: </strong>Our findings reveal a previously unrecognized MYCN-GAL-SREBP1 lipogenic axis that drives SCA proliferation. SREBP1-dependent lipogenesis represents a promising and druggable therapeutic vulnerability for the treatment of SCAs.</p>","PeriodicalId":19377,"journal":{"name":"Neuro-oncology","volume":" ","pages":"2401-2416"},"PeriodicalIF":13.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13635596/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148308859","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comprehensive performance testing and external validation of an AI algorithm to detect and segment brain metastases.","authors":"Rupesh Kotecha, Eyub Y Akdemir, Naseem Ud Din, Sreenija Yarlagadda, Mauricio Reyes, Francesco Nitti, Alonso N Gutierrez, D Jay M Wieczorek, Yongsook C Lee, Ranjini Tolakanahalli, Evan D Bander, Michael W McDermott, Manmeet S Ahluwalia, Minesh P Mehta","doi":"10.1093/neuonc/noag152","DOIUrl":"10.1093/neuonc/noag152","url":null,"abstract":"<p><strong>Background: </strong>Artificial intelligence (AI)-based models have shown initial promise in imaging brain metastasis; however, many lack validation against advanced imaging-informed datasets, precluding external validity and limiting widespread adoption. To overcome these limitations, we performed comprehensive performance testing against reference standard metrics and externally validated an AI algorithm.</p><p><strong>Methods: </strong>As part of its FDA-clearance process, performance testing of a previously developed U-Net-based AI model was conducted on a multi-institutional cohort with reference standard established via consensus review by 3 neuroradiologists. External validation was performed on patients imaged with dual sequences (augmented) as well as an open-access dataset (UCSF-BMSR). Evaluation metrics included sensitivity, false positive (FP) rate, positive predictive value (PPV), dice similarity coefficient (DSC), 95% Hausdorff distance, normalized surface distance, and qualitative physician assessment.</p><p><strong>Results: </strong>In the FDA performance testing cohort, the AI algorithm achieved a sensitivity of 90.0% (95% CI: 87.0%-94.0%), DSC of 0.86 (95% CI: 0.83-0.89), and average FP rate of 0.57 lesions. In the augmented and open-access external validation cohorts, a sensitivity of 81.4% (95% CI: 73.7%-89.1%) and 85.2% (95% CI: 83.0%-87.4%) with an average number of 0.22 and 1.19 FP lesions and DSCs of 0.70 (95% CI: 0.66-0.73) and 0.78 (95% CI: 0.77-0.78) were calculated, respectively. In the augmented external validation cohort, 38.5% of contours were rated as requiring major revisions.</p><p><strong>Conclusion: </strong>This AI algorithm demonstrated promising performance via 3 unique datasets. However, given the notable rate of contour revisions, these findings support its clinical role not as an autonomous system but as a human-in-the-loop tool requiring physician oversight.</p>","PeriodicalId":19377,"journal":{"name":"Neuro-oncology","volume":" ","pages":"2449-2460"},"PeriodicalIF":13.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13635595/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148412434","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}