Search

Background: Thyroid Transcription Factor-1 (TTF-1) expression in lung adenocarcinoma (LUAD) has been studied for its prognostic value in early-stage and metastatic disease. Its role in brain metastasis remains unexplored. This study investigates the predictive value and association of TTF-1 status with clinicopathological variables in patients with synchronous LUAD brain metastases., Material and Methods: In this bicentric retrospective study, 245 patients with newly diagnosed, treatment-naïve brain metastasis undergoing resection were included. Patient data were retrieved from electronic records. Outcomes included overall and progression-free survival. Statistical analysis included Kaplan-Meier estimates and Cox proportional hazards regression., Results: Mean Ki67 index in TTF-1 negative patients was 43% [95% CI 38-48%] compared to 32% [95% CI 29-35%] in TTF-1 positive (TTF-1 +) patients (p < 0.001). Tumor volume was significantly larger in TTF-1 negative (TTF-1-) patients (mean volume 24 mL [95% CI 18-31 mL]) vs. 15 mL [95% CI 12-17 mL] in TTF-1 + patients (padjust = 0.003). Perifocal edema was smaller in TTF-1- patients (mean volume: 58 mL [95% CI 45-70 mL]) vs. 84 mL [95% CI 73-94 mL] in TTF-1 + patients (padjust = 0.077). Tumor and edema volume did not correlate. TTF-1- patients showed worse overall, intracranial, and extracranial progression-free survival. In a multivariable Cox model, positive TTF-1 status was independently associated with improved outcomes. Negative TTF-1 status was associated with increased hazard for intracranial disease progression compared to extracranial progression., Conclusion: In synchronous LUAD brain metastases, TTF-1 negativity reflects an aggressive phenotype with larger proliferation capacity and tumor volume. Future research should explore the underlying cellular and molecular alterations of this phenotype., Competing Interests: Declarations. Competing interests: The authors declare no relevant financial or non-financial interests. There were no competing interests with respect to this study. Ethical approval and consent to participate: This study was approved by the local institutional review boards (EA1/399/20 and 20-1799-101). Consent for publication: Not applicable. Previous presentations: Portions of this work were presented at the annual meeting of the European Association of Neurosurgical Societies (EANS), Sofia, Bulgaria, October 13–17, 2024., (© 2024. The Author(s).)

The translation of AI-generated brain metastases (BM) segmentation into clinical practice relies heavily on diverse, high-quality annotated medical imaging datasets. The BraTS-METS 2023 challenge has gained momentum for testing and benchmarking algorithms using rigorously annotated internationally compiled real-world datasets. This study presents the results of the segmentation challenge and characterizes the challenging cases that impacted the performance of the winning algorithms. Untreated brain metastases on standard anatomic MRI sequences (T1, T2, FLAIR, T1PG) from eight contributed international datasets were annotated in stepwise method: published UNET algorithms, student, neuroradiologist, final approver neuroradiologist. Segmentations were ranked based on lesion-wise Dice and Hausdorff distance (HD95) scores. False positives (FP) and false negatives (FN) were rigorously penalized, receiving a score of 0 for Dice and a fixed penalty of 374 for HD95. The mean scores for the teams were calculated. Eight datasets comprising 1303 studies were annotated, with 402 studies (3076 lesions) released on Synapse as publicly available datasets to challenge competitors. Additionally, 31 studies (139 lesions) were held out for validation, and 59 studies (218 lesions) were used for testing. Segmentation accuracy was measured as rank across subjects, with the winning team achieving a LesionWise mean score of 7.9. The Dice score for the winning team was 0.65 ± 0.25. Common errors among the leading teams included false negatives for small lesions and misregistration of masks in space. The Dice scores and lesion detection rates of all algorithms diminished with decreasing tumor size, particularly for tumors smaller than 100 mm3. In conclusion, algorithms for BM segmentation require further refinement to balance high sensitivity in lesion detection with the minimization of false positives and negatives. The BraTS-METS 2023 challenge successfully curated well-annotated, diverse datasets and identified common errors, facilitating the translation of BM segmentation across varied clinical environments and providing personalized volumetric reports to patients undergoing BM treatment., Competing Interests: Conflicts of Interest No conflicts of interest to disclose.

Background: Radionecrosis is a common complication in radiation oncology, while mechanisms and risk factors have yet to be fully explored. We therefore conducted a systematic review to understand the pathogenesis and identify factors that significantly affect the development., Methods: We performed a systematic literature search based on the PRISMA guidelines using PubMed, Ovid, and Web of Science databases. The complete search strategy can be found as a preregistered protocol on PROSPERO (CRD42023361662)., Results: We included 83 studies, most involving healthy animals (n = 72, 86.75 %). High doses of hemispherical irradiation of 30 Gy in rats and 50 Gy in mice led repeatedly to radionecrosis among different studies and set-ups. Higher dose and larger irradiated volume were associated with earlier onset. Fractionated schedules showed limited effectiveness in the prevention of radionecrosis. Distinct anatomical brain structures respond to irradiation in various ways. White matter appears to be more vulnerable than gray matter. Younger age, more evolved animal species, and genetic background were also significant factors, whereas sex was irrelevant. Only 13.25 % of the studies were performed on primary brain tumor bearing animals, no studies on brain metastases are currently available., Conclusion: This systematic review identified various factors that significantly affect the induction of radionecrosis. The current state of research neglects the utilization of animal models of brain tumors, even though patients with brain malignancies constitute the largest group receiving brain irradiation. This latter aspect should be primarily addressed when developing an experimental radionecrosis model for translational implementation., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 The Authors. Published by Elsevier B.V. All rights reserved.)

Introduction: Magnetic Resonance Elastography (MRE) allows the non-invasive quantification of tumor biomechanical properties in vivo . With increasing incidence of brain metastases, there is a notable absence of appropriate preclinical models to investigate their biomechanical characteristics. Therefore, the purpose of this work was to assess the biomechanical characteristics of B16 melanoma brain metastases (MBM) and compare it to murine GL261 glioblastoma (GBM) model using multifrequency MRE with tomoelastography post processing., Methods: Intracranial B16 MBM (n = 6) and GL261 GBM (n = 7) mouse models were used. Magnetic Resonance Imaging (MRI) was performed at set intervals after tumor implantation: 5, 7, 12, 14 days for MBM and 13 and 22 days for GBM. The investigations were performed using a 7T preclinical MRI with 20 mm head coil. The protocol consisted of single-shot spin echo-planar multifrequency MRE with tomoelastography post processing, contrast-enhanced T1- and T2-weighted imaging and diffusion-weighted imaging (DWI) with quantification of apparent diffusion coefficient of water (ADC). Elastography quantified shear wave speed (SWS), magnitude of complex MR signal (T2/T2*) and loss angle (φ). Immunohistological investigations were performed to assess vascularization, blood-brain-barrier integrity and extent of glucosaminoglucan coverage., Results: Volumetric analyses displayed rapid growth of both tumor entities and softer tissue properties than healthy brain (healthy: 5.17 ± 0.48, MBM: 3.83 ± 0.55, GBM: 3.7 ± 0.23, [m/s]). SWS of MBM remained unchanged throughout tumor progression with decreased T2/T2* intensity and increased ADC on days 12 and 14 (p<0.0001 for both). Conversely, GBM presented reduced φ values on day 22 (p=0.0237), with no significant alterations in ADC. Histological analysis revealed substantial vascularization and elevated glycosaminoglycan content in both tumor types compared to healthy contralateral brain., Discussion: Our results indicate that while both, MBM and GBM, exhibited softer properties compared to healthy brain, imaging and histological analysis revealed different underlying microstructural causes: hemorrhages in MBM and increased vascularization and glycosaminoglycan content in GBM, further corroborated by DWI and T2/T2* contrast. These findings underscore the complementary nature of MRE and its potential to enhance our understanding of tumor characteristics when used alongside established techniques. This comprehensive approach could lead to improved clinical outcomes and a deeper understanding of brain tumor pathophysiology., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Janas, Jordan, Bertalan, Meyer, Bukatz, Sack, Senger, Nieminen-Kelhä, Brandenburg, Kremenskaia, Krantchev, Al-Rubaiey, Mueller, Koch, Boehm-Sturm, Reiter, Zips, Vajkoczy and Acker.)

Background: Emerging evidence suggests that treatment of NSCLC brain metastases with immune checkpoint inhibitors (ICIs) is associated with response rates similar to those of extracranial disease. Programmed death-ligand 1 (PD-L1) tumor proportion score (TPS) serves as a predictive biomarker for ICI response. However, the predictive value of brain metastasis-specific (intracranial) PD-L1 TPS is not established. We investigated the role of intra- and extracranial PD-L1 TPS in NSCLC patients treated with ICI following brain metastasis resection., Methods: Clinical data from NSCLC patients treated with ICI following brain metastasis resection (n = 64) were analyzed. PD-L1 TPS of brain metastases (n = 64) and available matched extracranial tumor tissue (n = 44) were assessed via immunohistochemistry. Statistical analyses included cut point estimation via maximally selected rank statistics, Kaplan-Meier estimates, and multivariable Cox regression analysis for intracranial progression-free survival (icPFS), extracranial progression-free survival (ecPFS), and overall survival (OS)., Results: PD-L1 expression was found in 54.7% of brain metastases and 68.2% of extracranial tumor tissues, with a median intra- and extracranial PD-L1 TPS of 7.5% (0 - 50%, IQR) and 15.0% (0 - 80%, IQR), respectively. In matched tissue samples, extracranial PD-L1 TPS was significantly higher than intracranial PD-L1 TPS (p = 0.013). Optimal cut points for intracranial and extracranial PD-L1 TPS varied according to outcome parameter assessed. Notably, patients with a high intracranial PD-L1 TPS (> 40%) exhibited significantly longer icPFS as compared to patients with a low intracranial PD-L1 TPS (≤ 40%). The cut point of 40% for intracranial PD-L1 TPS was independently associated with OS, icPFS and ecPFS in multivariable analyses., Conclusion: Our study highlights the potential role of intracranial PD-L1 TPS in NSCLC, which could be used to predict ICI response in cases where extracranial tissue is not available for PD-L1 assessment as well as to specifically predict intracranial response., (© 2024. The Author(s).)