To assess the feasibility of using hypoxia-powered magnetic resonance imaging (MRI) to guide prospective sampling of high-response habitats beyond contrast enhancement in glioblastoma. Nine treatment-naive patients with histologically confirmed isocitrate dehydrogenase (IDH)-wildtype glioblastoma underwent blood oxygenation level-dependent (BOLD) MRI during a controlled transient hypoxic stimulus on the evening before surgery. Anatomical segmentation and voxel-wise temporal decomposition generated normalized hypoxia response maps. Otsu thresholding defined a high-response habitat outside contrast enhancement. Targets were confirmed with the surgeon according to anatomical accessibility and eloquence, transferred to neuronavigation, and sampled before major debulking. We divided the specimen for formalin-fixed, paraffin-embedded neuropathology and frozen-tissue storage. Neuronavigation transfer succeeded in all nine cases. Eight targets were accessible and sampled. Seven sampled targets contained tumor. One yielded reactive cortex and white matter without definite tumor. No target-sampling complications were observed among the eight sampled procedures. Hypoxia-powered MRI could be integrated into prospective neuronavigation-guided sampling beyond contrast enhancement. Tumor was identified in seven of eight sampled targets. By translating a tracer-free physiological response map into coordinate-matched tissue, the GLIOXY workflow enables prospective, targeted tissue sampling of non-enhancing glioblastoma regions.
Intracranial germinomas are highly radiosensitive tumors associated with excellent long-term outcomes. Whole-ventricle irradiation (WVI) with a tumor-bed boost, with or without chemotherapy, has become a standard treatment for localized disease, balancing disease control with reduced treatment-related toxicity. This retrospective study included patients with pure intracranial germinoma treated at Children's Cancer Hospital Egypt (CCHE) between 2008 and 2024. Patients received WVI to 23.4 Gy followed by a tumor-bed boost up to a total dose of 45 Gy. The objectives were to characterize relapse patterns and describe survival outcomes. A total of 63 patients met the eligibility criteria and were included in the analysis. Median follow-up was 62.9 months. Six patients (9.5
Glioblastoma (GBM) exhibits profound therapeutic resistance that is largely attributed to its remarkable cellular plasticity and capacity to adapt to therapeutic and environmental stress. Although suppression of mechanistic target of rapamycin (mTOR) signaling has been extensively investigated as a therapeutic strategy, the adaptive mechanisms that limit its efficacy remain incompletely understood. Human GBM cells were subjected to sphere-forming conditions or pharmacological inhibition of mTOR signaling to evaluate stress-induced cellular responses. Proteomic profiling following mTORC2 inhibition was performed to identify adaptive metabolic alterations. Functional analyses were conducted using NFS1 cysteine desulfurase (NFS1) knockdown, orthotopic xenograft models, and publicly available datasets. Sphere-forming conditions and pharmacological inhibition of mTOR signaling induced stemness-associated changes in GBM cells. Unbiased proteomic profiling identified NFS1 as a stress-responsive protein induced by mTOR suppression. NFS1 depletion attenuated selected stemness-associated responses and sensitized GBM cells to mTOR pathway inhibition. Increased NFS1 expression was further confirmed in highly tumorigenic orthotopic xenografts, whereas exploratory analyses suggested a potential modulatory role of iron availability in this adaptive phenotype. Our findings identify NFS1 as a stress-induced metabolic dependency that emerges during adaptation to mTOR suppression in GBM. Targeting adaptive vulnerabilities by therapeutic stress, rather than constitutive oncogenic pathways alone, may represent a promising strategy for overcoming treatment resistance in malignant gliomas.
Combining stereotactic radiosurgery (SRS) with systemic therapy is increasingly common for brain metastases, yet real-world evidence on immunotherapy timing relative to SRS remains limited. We evaluated the association of immunotherapy and its timing with survival, intracranial control and toxicity, with exploratory analyses of chemotherapy and targeted therapy. We retrospectively analyzed 155 patients with 556 brain metastases treated with SRS at a single tertiary center (January 2021–February 2025). Patient-level associations of systemic therapy, overall and by timing (before, concurrent with or after SRS), with overall survival, 1-year local, distant intracranial and systemic progression, toxicity and radionecrosis were assessed using Kaplan–Meier, Cox and logistic regression analyses, with false discovery rate correction for immunotherapy timing. Median follow-up was 18.8 months. Immunotherapy (n = 88) was not associated with overall survival (adjusted HR 0.79, 95
Local control after Gamma Knife radiosurgery (GKRS) for radioresistant brain metastases remains variable. Melanoma and renal cell carcinoma brain metastases are relatively radioresistant, yet pre-treatment prediction of local failure and its timing remains limited. We developed and internally validated a survival machine learning model to predict lesion-specific local failure and time to local failure using only pre-GKRS variables. We performed a retrospective single-center study of melanoma and renal cell carcinoma brain metastases treated with GKRS. Local failure was modeled as a right-censored survival outcome. A Random Survival Forest (RSF) was trained using baseline demographic, clinical, and tumor-level variables available before treatment. The individual lesion was the prediction unit, while data partitioning was performed at the patient level to minimize information leakage. Performance was assessed using the concordance index, cumulative/dynamic AUC, and Brier score with patient-grouped 5-fold internal cross-validation. The cohort included 893 lesions, of which 228 (25.5
