New insights into neuropathology and pathogenesis of autoimmune glial fibrillary acidic protein meningoencephalomyelitis.

Anti-glial fibrillary acidic protein (GFAP) meningoencephalomyelitis (autoimmune GFAP astrocytopathy) is a new autoimmune central nervous system (CNS) disease diagnosable by the presence of anti-GFAP autoantibodies in the cerebrospinal fluid and presents as meningoencephalomyelitis in the majority of patients. Only few neuropathological reports are available and little is known about the pathogenic mechanisms. We performed a histopathological study of two autopsies and nine CNS biopsies of patients with anti-GFAP autoantibodies and found predominantly a lymphocytic and in one autopsy case a granulomatous inflammatory phenotype. Inflammatory infiltrates were composed of B and T cells, including tissue-resident memory T cells. Although obvious astrocytic damage was absent in the GFAP-staining, we found cytotoxic T cell-mediated reactions reflected by the presence of CD8+/perforin+/granzyme A/B+ cells, polarized towards astrocytes. MHC-class-I was upregulated in reactive astrocytes of all biopsies and two autopsies but not in healthy controls. Importantly, we observed a prominent immunoreactivity of astrocytes with the complement factor C4d. Finally, we provided insight into an early phase of GFAP autoimmunity in an autopsy of a pug dog encephalitis that was characterized by marked meningoencephalitis with selective astrocytic damage with loss of GFAP and AQP4 in the lesions.Our histopathological findings indicate that a cytotoxic T cell-mediated immune reaction is present in GFAP autoimmunity. Complement C4d deposition on astrocytes could either represent the cause or consequence of astrocytic reactivity. Selective astrocytic damage is prominent in the early phase of GFAP autoimmunity in a canine autopsy case, but mild or absent in subacute and chronic stages in human disease, probably due to the high regeneration potential of astrocytes. The lymphocytic and granulomatous phenotypes might reflect different stages of lesion development or patient-specific modifications of the immune response. Future studies will be necessary to investigate possible implications of pathological subtypes for clinical disease course and therapeutic strategies.

Anti-Ri paraneoplastic neurological syndrome presenting with bilateral cranial nerve VI palsy and jaw dystonia-a distinctive syndrome within the anti-Ri spectrum? : Case report and literature review.

OBJECTIVE: Paraneoplastic neurological syndromes (PNS) are rare disorders associated with various onconeuronal antibodies. Anti-Ri antibodies (ANNA-2) are typically found in patients with opsoclonus myoclonus syndrome (OMS) and ataxia. CASE REPORT: We present an anti-Ri antibody-positive 77-year-old woman with subacute progressive bilateral cranial nerve VI palsy, gait disturbance and jaw dystonia. MRI of the brain showed hyperintense signals on T2 bitemporal without contrast enhancement. Cerebrospinal fluid (CSF) examination exhibited mild pleocytosis of 13 cells/µl and positive oligoclonal bands. CSF was overall inconspicuous for a malignant or inflammatory etiology. Immunofluorescence analysis revealed anti-Ri antibodies in both serum and CSF. Subsequent diagnostic work up resulted in a newly diagnosed ductal carcinoma of the right breast. PNS in this case partially responded to the anti-tumor therapy. CONCLUSION: This case shows similarities with recently published anti-Ri syndromes, which might form a distinct triad within the anti-Ri spectrum.

Perioperative levetiracetam for seizure prophylaxis in seizure-naive brain tumor patients with focus on neurocognitive functioning.

INTRODUCTION: In seizure-naive brain tumor patients, the efficacy of perioperative prophylactic antiepileptic drug treatment remains controversial. In case of administration, the common preferred drug is levetiracetam (LEV) because of its favorable pharmacological profile. Research to date has not sufficiently determined how LEV affects cognition in the short term, as is the case in the perioperative period. The objective of this prospective study was to examine the neurocognitive functioning of seizure-naive brain tumor patients after receiving LEV perioperatively. METHODS: Fortythree patients with supratentorial brain tumor scheduled for surgery received LEV three days before until six days after surgery as seizure prophylaxis. Cognitive functioning (NeuroCogFX), LEV plasma-levels, hematotoxicity, side-effects, as well as health-related quality of life (HRQoL, Qolie31), were recorded preoperatively before (Baseline) and after onset of LEV (Pre-Op), 4-6 days postoperatively (Post-Op) and 21 days postoperatively (Follow-Up). RESULTS: No significant changes in cognitive functioning and HRQoL were seen after onset of preoperative LEV. There was a significant improvement of NeuroCogFX total-score at Follow-Up (p = 0.004) compared to Baseline. The overall-score Qolie31 showed simultaneous improvement patterns as cognitive functioning (p < 0.001). The most frequent side effect related to study drug was somnolence (in 28.6% of patients). CONCLUSIONS: A significant improvement of cognitive functioning, as well as an improvement in HRQoL, were detected postoperatively. This is presumably due to the debulking effect of the surgery. Nevertheless, LEV has no detrimental effect on cognitive functioning in the perioperative phase in seizure-naive brain tumor patients. TRIAL REGISTRATION: This study was registered prospectively (Date: 25/11/2015; EudraCT: 2015-003,916-19).

Physiological MRI of microvascular architecture, neovascularization activity, and oxygen metabolism facilitate early recurrence detection in patients with IDH-mutant WHO grade 3 glioma.

PURPOSE: This study aimed to determine the diagnostic performance of physiological MRI biomarkers including microvascular perfusion and architecture, neovascularization activity, tissue oxygen metabolism, and tension for recurrence detection of IDH-mutant WHO grade 3 glioma. METHODS: Sixty patients with IDH-mutant WHO grade 3 glioma who received overall 288 follow-up MRI examinations at 3 Tesla after standard treatment were retrospectively evaluated. A conventional MRI protocol was extended with a physiological MRI approach including vascular architecture mapping and quantitative blood-oxygen-level-dependent imaging which required 7 min extra data acquisition time. Custom-made MATLAB software was used for the calculation of MRI biomarker maps of microvascular perfusion and architecture, neovascularization activity, tissue oxygen metabolism, and tension. Statistical procedures included receiver operating characteristic analysis. RESULTS: Overall, 34 patients showed recurrence of the WHO grade 3 glioma; of these, in 15 patients, recurrence was detected one follow-up examination (averaged 160 days) earlier by physiological MRI data than by conventional MRI. During this time period, the tumor volume increased significantly (P = 0.001) on average 7.4-fold from 1.5 to 11.1 cm3. Quantitative analysis of MRI biomarkers demonstrated microvascular but no macrovascular hyperperfusion in early recurrence. Neovascularization activity (AUC = 0.833), microvascular perfusion (0.682), and oxygen metabolism (0.661) showed higher diagnostic performance for early recurrence detection of WHO grade 3 glioma compared to conventional MRI including cerebral blood volume (0.649). CONCLUSION: This study demonstrated that the targeted assessment of microvascular features and tissue oxygen tension as an early sign of neovascularization activity provided valuable information for recurrence diagnostic of WHO grade 3 glioma.

Influence of preoperative corticosteroid treatment on rate of diagnostic surgeries in primary central nervous system lymphoma: a multicenter retrospective study.

BACKGROUND: Corticosteroid therapy (CST) prior to biopsy may hinder histopathological diagnosis in primary central nervous system lymphoma (PCNSL). Therefore, preoperative CST in patients with suspected PCNSL should be avoided if clinically possible. The aim of this study was thus to analyze the difference in the rate of diagnostic surgeries in PCNSL patients with and without preoperative CST. METHODS: A multicenter retrospective study including all immunocompetent patients diagnosed with PCNSL between 1/2004 and 9/2018 at four neurosurgical centers in Austria was conducted and the results were compared to literature. RESULTS: A total of 143 patients were included in this study. All patients showed visible contrast enhancement on preoperative MRI. There was no statistically significant difference in the rate of diagnostic surgeries with and without preoperative CST with 97.1% (68/70) and 97.3% (71/73), respectively (p = 1.0). Tapering and pause of CST did not influence the diagnostic rate. Including our study, there are 788 PCNSL patients described in literature with an odds ratio for inconclusive surgeries after CST of 3.3 (CI 1.7-6.4). CONCLUSIONS: Preoperative CST should be avoided as it seems to diminish the diagnostic rate of biopsy in PCNSL patients. Yet, if CST has been administered preoperatively and there is still a contrast enhancing lesion to target for biopsy, surgeons should try to keep the diagnostic delay to a minimum as the likelihood for acquiring diagnostic tissue seems sufficiently high.

Clinical characteristics and prognostic factors of adult patients with pilocytic astrocytoma.

INTRODUCTION: Pilocytic astrocytoma (PA) is the most common primary brain neoplasm in children and treated in curative intent with gross total resection (GTR). However, PA is rare in adults, resulting in limited knowledge on the natural clinical course. This study aimed to describe the clinical course and identify prognostic factors of adult patients with PA. METHODS: 46 patients ≥ 18 years at diagnosis of PA and neurosurgical resection or biopsy between 2000 and 2018 were identified from the Neuro-Biobank of the Medical University of Vienna. In two cases with differing histopathological diagnosis at recurrence, DNA methylation analysis was performed using Illumina Infinium HumanMethylation850 BeadChip (850 k) arrays and the Molecular Neuropathology classifier. Clinico-pathological features were correlated with patient outcomes. RESULTS: Median age at diagnosis was 32.5 years (range: 19-75) and median Ki67 proliferation index was 2.8% (0.5-13.4%). Tumor location significantly correlated with resectability (p < 0.001). Tumor progression or recurrence was observed in 9/46 (19.6%) patients after a median follow up time of 53.0 months (range 0.5-300). 5-year overall and progression-free survival rates were 85.3% and 70.0%, respectively. 2/9 (22.2%) patients presented with histological changes in the recurrent tumor specimen. In detail, methylation classification redefined the histological diagnosis to anaplastic astrocytoma with piloid features and glioma in one patient, each. Age > 40 and higher body mass index (BMI) were associated with impaired progression-free and overall survival (p < 0.05). CONCLUSIONS: Tumor recurrence or progression in adult PA patients was higher than the one reported in pediatric patients. Higher age and BMI were associated with impaired prognosis.

Palliative care in glioma management.

PURPOSE OF REVIEW: The concept of palliative care is becoming increasingly important in the management of glioma patients. However, the right time for implementation, as well as the individual degree of integration of palliative care aspects, are still a matter of debate. This review updates recent evidence of palliative care in glioma and raises questions for future developments. RECENT FINDINGS: According to the disease-specific aspects of palliative care in glioma management, there is an increasing need for a specialized 'neurooncological' palliative care approach.The implementation of palliative care for glioma patients and caregivers avoids unnecessary hospitalization and reduces health-related costs. Moreover, palliative care may be offered in different settings (inpatient/outpatient) according to local health structures, but definitely improves the QOL of glioma patients and their caregivers. SUMMARY: There are considerable differences between countries with respect to palliative care in glioma. Major aspects of good-quality care throughout the countries are optimized symptom control, support for patients and proxies by an interdisciplinary team and to provide realistic information to patients and caregivers. The availability of neurooncological palliative teams and multidisciplinary support programs increases well being of glioma patients and caregivers as well as releases pressure on healthcare systems.

Increasing use of immunotherapy and prolonged survival among younger patients with primary CNS lymphoma: a population-based study.

Background: Primary CNS lymphoma is a highly aggressive and rare type of extranodal non-Hodgkin lymphoma. Although, new therapeutic approaches have led to improved survival, the management of the disease poses a challenge, practice patterns vary across institutions and countries, and remain ill-defined for vulnerable patient subgroups. Material and Methods: Using information from the Austrian Brain Tumor Registry we followed a population-based cohort of 189 patients newly diagnosed from 2005 to 2010 through various lines of treatment until death or last follow-up (12-31-2016). Prognostic factors and treatment-related data were integrated in a comprehensive survival analysis including conditional survival estimates. Results: We find variable patterns of first-line treatment with increasing use of rituximab and high-dose methotrexate (HDMTX)-based poly-chemotherapy after 2007, paralleled by an increase in median overall survival restricted to patients aged below 70 years. In the entire cohort, 5-year overall survival was 24.4% while 5-year conditional survival increased with every year postdiagnosis. Conclusion: In conclusion, we show that the use of poly-chemotherapy and immunotherapy has disseminated to community practice to a fair extent and survival has increased over time at least in younger patients. Annually increasing conditional survival rates provide clinicians with an adequate and encouraging prognostic measure.

Vascular architecture mapping for early detection of glioblastoma recurrence.

OBJECTIVE: Treatment failure and inevitable tumor recurrence are the main reasons for the poor prognosis of glioblastoma (GB). Gross-total resection at repeat craniotomy for GB recurrence improves patient overall survival but requires early and reliable detection. It is known, however, that even advanced MRI approaches have limited diagnostic performance for distinguishing tumor progression from pseudoprogression. The novel MRI technique of vascular architectural mapping (VAM) provides deeper insight into tumor microvascularity and neovascularization. In this study the authors evaluated the usefulness of VAM for the monitoring of GB patients and quantitatively analyzed the features of neovascularization of early- and progressed-stage GB recurrence. METHODS: In total, a group of 115 GB patients who received overall 374 follow-up MRI examinations after standard treatment were retrospectively evaluated in this study. The clinical routine MRI (cMRI) protocol at 3 Tesla was extended with the authors' experimental VAM approach, requiring 2 minutes of extra time for data acquisition. Custom-made MATLAB software was used for calculation of imaging biomarker maps of macrovascular perfusion from perfusion cMRI as well as of microvascular perfusion and architecture from VAM data. Additionally, cMRI data were analyzed by two board-certified radiologists in consensus. Statistical procedures included receiver operating characteristic (ROC) analysis to determine diagnostic performances for GB recurrence detection. RESULTS: Overall, cMRI showed GB recurrence in 89 patients, and in 28 of these patients recurrence was detected earlier with VAM data, by 1 (20 patients) or 2 (8 patients) follow-up examinations, than with cMRI data. The mean time difference between recurrence detection with VAM and cMRI data was 147 days. During this time period the mean tumor volume increased significantly (p < 0.001) from 9.7 to 26.8 cm3. Quantitative analysis of imaging biomarkers demonstrated microvascular but no macrovascular hyperperfusion in early GB recurrence. Therefore, ROC analysis revealed superior diagnostic performance for VAM compared with cMRI. CONCLUSIONS: This study demonstrated that the targeted assessment of microvascular features using the VAM technique provided valuable information about early neovascularization activity in recurrent GB that is complementary to perfusion cMRI and may be helpful for earlier and more precise monitoring of patients suffering from GB. This VAM approach is compatible with existing cMRI protocols. Prospective clinical trials are necessary to investigate the clinical usefulness and potential benefit of increased overall survival with the use of VAM in patients with recurrent GB.

European Association for Neuro-Oncology (EANO) guidelines for palliative care in adults with glioma.

Patients with glioma present with complex palliative care needs throughout their disease trajectory. The life-limiting nature of gliomas and the presence of specific symptoms related to neurological deterioration necessitate an appropriate and early palliative care approach. The multidisciplinary palliative care task force of the European Association of Neuro-Oncology did a systematic review of the available scientific literature to formulate the best possible evidence-based recommendations for the palliative care of adult patients with glioma, with the aim to reduce symptom burden and improve the quality of life of patients and their caregivers, particularly in the end-of-life phase. When recommendations could not be made because of the scarcity of evidence, the task force either used evidence from studies of patients with systemic cancer or formulated expert opinion. Areas of palliative care that currently lack evidence and thus deserve attention for further research are fatigue, disorders of behaviour and mood, interventions for the needs of caregivers, and timing of advance care planning.

MR Imaging-derived Oxygen Metabolism and Neovascularization Characterization for Grading and IDH Gene Mutation Detection of Gliomas.

Purpose To explore the diagnostic performance of physiological magnetic resonance (MR) imaging of oxygen metabolism and neovascularization activity for grading and characterization of isocitrate dehydrogenase (IDH) gene mutation status of gliomas. Materials and Methods This retrospective study had institutional review board approval; written informed consent was obtained from all patients. Eighty-three patients with histopathologically proven glioma (World Health Organization [WHO] grade II-IV) were examined with quantitative blood oxygen level-dependent imaging and vascular architecture mapping. Biomarker maps of neovascularization activity (microvessel radius, microvessel density, and microvessel type indicator [MTI]) and oxygen metabolism (oxygen extraction fraction [OEF] and cerebral metabolic rate of oxygen [CMRO2]) were calculated. Receiver operating characteristic analysis was used to determine diagnostic performance for grading and detection of IDH gene mutation status. Results Low-grade (WHO grade II) glioma showed areas with increased OEF (+18%, P < .001, n = 20), whereas anaplastic glioma (WHO grade III) and glioblastoma (WHO grade IV) showed decreased OEF when compared with normal brain tissue (-54% [P < .001, n = 21] and -49% [P < .001, n = 41], respectively). This allowed clear differentiation between low- and high-grade glioma (area under the receiver operating characteristic curve [AUC], 1) for the patient cohort. MTI had the highest diagnostic performance (AUC, 0.782) for differentiation between gliomas of grades III and IV among all biomarkers. CMRO2 was decreased (P = .037) in low-grade glioma with a mutated IDH gene, and MTI was significantly increased in glioma grade III with IDH mutation (P = .013) when compared with the IDH wild-type counterparts. CMRO2 showed the highest diagnostic performance for IDH gene mutation detection in low-grade glioma (AUC, 0.818) and MTI in high-grade glioma (AUC, 0.854) and for all WHO grades (AUC, 0.899) among all biomarkers. Conclusion MR imaging-derived oxygen metabolism and neovascularization characterization may be useful for grading and IDH mutation detection of gliomas and requires only 7 minutes of extra imaging time. © RSNA, 2016 Online supplemental material is available for this article.

Machine Learning-Based Prediction of Glioma IDH Gene Mutation Status Using Physio-Metabolic MRI of Oxygen Metabolism and Neovascularization (A Bicenter Study).

The mutational status of the isocitrate dehydrogenase (IDH) gene plays a key role in the treatment of glioma patients because it is known to affect energy metabolism pathways relevant to glioma. Physio-metabolic magnetic resonance imaging (MRI) enables the non-invasive analysis of oxygen metabolism and tissue hypoxia as well as associated neovascularization and microvascular architecture. However, evaluating such complex neuroimaging data requires computational support. Traditional machine learning algorithms and simple deep learning models were trained with radiomic features from clinical MRI (cMRI) or physio-metabolic MRI data. A total of 215 patients (first center: 166 participants + 16 participants for independent internal testing of the algorithms versus second site: 33 participants for independent external testing) were enrolled using two different physio-metabolic MRI protocols. The algorithms trained with physio-metabolic data demonstrated the best classification performance in independent internal testing: precision, 91.7%; accuracy, 87.5%; area under the receiver operating curve (AUROC), 0.979. In external testing, traditional machine learning models trained with cMRI data exhibited the best IDH classification results: precision, 84.9%; accuracy, 81.8%; and AUROC, 0.879. The poor performance for the physio-metabolic MRI approach appears to be explainable by site-dependent differences in data acquisition methodologies. The physio-metabolic MRI approach potentially supports reliable classification of IDH gene status in the presurgical stage of glioma patients. However, non-standardized protocols limit the level of evidence and underlie the need for a reproducible framework of data acquisition techniques.

CXCL13 as a biomarker in the diagnostics of European lyme Neuroborreliosis - A prospective multicentre study in Austria.

Background: 'Definite Neuroborreliosis (NB)' is diagnosed with the presence of NB-specific symptoms, cerebrospinal fluid (CSF) pleocytosis and an elevated Borrelia Burgdorferi antibody index. However, some diagnostic uncertainties exist. The B-cell chemokine CXCL13 represents an emerging biomarker for the diagnosis and treatment of NB because its intrathecal concentration rises prior to the Borrelia antibody index and drops rapidly after antibiotic therapy. Nevertheless, due to lacking prospective data, a definite CXCL13 cut-off for the diagnosis of NB is still pending. Objective: Definition of a CSF CXCL13 cut-off for the diagnosis of acute and untreated NB in a prospective study setting. Design and methods: This multicentre prospective study involved 6 neurological departments treating patients in the Lower Austria district (1.7 million inhabitants). The controls were patients scheduled for a spinal tap but not clinically diagnosed with NB. Demographic data, clinical characteristics and blood counts, as well as inflammatory CSF values and CSF CXCL13-concentration were analysed. Results: We recruited 440 adult patients, of whom 42 have been diagnosed as having an acute and untreated 'definite NB'. Three hundred ninety-eight patients were assigned to the control group. The median intrathecal CXCL13 concentration was 2384 pg/ml for patients with NB and 0 pg/ml for controls. The difference was highly statistically significant (P ≤ .001). A CSF CXCL13 cut-off of 271 pg/ml resulted in a sensitivity of 95.2% and a specificity of 97.2% for the confirmation or exclusion of NB. Conclusion: Based on our results, we propose a CSF CXCL13 cut-off of 271 pg/ml with Euroimmun-Elisa for the diagnosis of acute and untreated NB. Due to its high sensitivity and specificity, CXCL13 is a strong candidate biomarker for routine NB assessment, especially in clinically unclear cases.

The caregivers' perspective on the end-of-life phase of glioblastoma patients.

Glioblastoma multiforme (GBM) still harbors a fatal prognosis. The involvement of the neurocognition and psyche poses unique challenges for care provision by relatives. We lack data about the caregivers' perspective on the end-of-life (EOL) phase of GBM patients to improve counseling and support. In this study we investigated the experiences of 52 caregivers of deceased GBM patients treated in Austria. We used a questionnaire developed by the University Medical Centre of Amsterdam for exploration of the EOL-phase in glioma patients. The caregivers (17 men, 34 women) completed the questionnaire in median three years after the patients' death. 29 % of caregivers reported that they felt incompletely prepared for their tasks, however, those with higher education levels felt significantly better informed. 29 % suffered from financial difficulties, which was associated with burnout (60 %) and reduced quality of life (QOL). The patients' most common symptoms reported by caregivers were fatigue (87 %), reduced consciousness (81 %) and aphasia (77 %). 22 % of patients were bedbound during their last three months increasing to 80 % in the last week of life. The reported QOL of caregivers was very low and did not differ between caregivers of patients, who died at home (40 %) and caregivers of patients, who died in hospital (46 %). The caregiver reported that their QOL was only slightly better than the QOL they attributed to the patients. Furthermore, the high frequency of financial difficulties, burnout symptoms and feelings of insufficient information emphasize the urgent need for support and training dedicated to caregivers.

Clinical neuropathology practice guide 1-2013: molecular subtyping of glioblastoma: ready for clinical use?

Recently, integrated genomewide analyses have revealed several glioblastoma (GB) subtypes, which differ in terms of key pathogenetic pathways and point to different cells of origin. Even though the proneural and mesenchymal GB signatures evolved as most robust, there is no consensus on the exact number of subtypes and defining criteria. Moreover, important issues concerning within-tumor heterogeneity and class-switching upon recurrence remain to be addressed. Early evidence indicates an association of different GB subtypes with patient outcome and response to therapy, which argues for the implementation of molecular GB subtyping, and consideration of GB subtypes in subsequent patient management. As genome-wide analyses are not routinely available to the majority of neuropathology laboratories, first attempts to implement immunohistochemical testing of surrogate markers are underway. However, so far, confirmatory studies are lacking and there is no consensus on which markers to use. Further, the rationale for testing is compromised from a clinical point of view by a lack of effective therapies for individual GB subtypes. Thus, incorporation of genomic research findings as a basis for GB patient management and clinical decision making currently remains a perspective for the future.

Radiomic features define risk and are linked to DNA methylation attributes in primary CNS lymphoma.

Background: The prognostic roles of clinical and laboratory markers have been exploited to model risk in patients with primary CNS lymphoma, but these approaches do not fully explain the observed variation in outcome. To date, neuroimaging or molecular information is not used. The aim of this study was to determine the utility of radiomic features to capture clinically relevant phenotypes, and to link those to molecular profiles for enhanced risk stratification. Methods: In this retrospective study, we investigated 133 patients across 9 sites in Austria (2005-2018) and an external validation site in South Korea (44 patients, 2013-2016). We used T1-weighted contrast-enhanced MRI and an L1-norm regularized Cox proportional hazard model to derive a radiomic risk score. We integrated radiomic features with DNA methylation profiles using machine learning-based prediction, and validated the most relevant biological associations in tissues and cell lines. Results: The radiomic risk score, consisting of 20 mostly textural features, was a strong and independent predictor of survival (multivariate hazard ratio = 6.56 [3.64-11.81]) that remained valid in the external validation cohort. Radiomic features captured gene regulatory differences such as in BCL6 binding activity, which was put forth as testable treatment target for a subset of patients. Conclusions: The radiomic risk score was a robust and complementary predictor of survival and reflected characteristics in underlying DNA methylation patterns. Leveraging imaging phenotypes to assess risk and inform epigenetic treatment targets provides a concept on which to advance prognostic modeling and precision therapy for this aggressive cancer.

Malignant spinal cord compression in cerebral glioblastoma multiforme: a multicenter case series and review of the literature.

Glioblastoma multiforme (GBM) is the most common malignant primary brain tumor in adults. Compared with other malignancies, remote metastases in GBM are rare. However, multicentric spreading within the central nervous system is common and also metastases to the spinal cord have been reported. Some of these drop metastases may also lead to malignant spinal cord compression (MSCC). We retrospectively identified nine patients from 2001 to 2010 and performed data analysis according to a standardized clinical protocol. We also provide a review of the literature on this rare condition. MSCC from cerebral GBM is rare and is found in approximately 1 % of GBM patients. Median age of 54 years in this case series is comparable with that of GBM patients without MSCC. Treatment regimens for cerebral GBM and overall survival was similar to those for patients without MSCC. Spinal metastasis seems to occur in the advanced state of the disease, and the outcome subsequently is extremely poor. All patients presented with multicentric radiological features of GBM on cerebral MRI when MSCC was diagnosed. Subependymal enhancement is another common radiological finding in GBM patients with spinal drop metastases. Steroids and focal radiotherapy were used to treat all patients, with little clinical benefit. This study is the largest case series of MSCC from cerebral GBM. Multicentric cerebral distribution and subependymal enhancement of GBM are observed on cerebral MRI at the time of MSCC. On the basis of our results, no specific treatment recommendations for MSCC in GBM patients can be given. However, accurate diagnosis of MSCC in GBM patients with spinal signs and symptoms can lead to adequate management of symptoms and improvement of quality of life in terms of best palliative care.

Radiophysiomics: Brain Tumors Classification by Machine Learning and Physiological MRI Data.

The precise initial characterization of contrast-enhancing brain tumors has significant consequences for clinical outcomes. Various novel neuroimaging methods have been developed to increase the specificity of conventional magnetic resonance imaging (cMRI) but also the increased complexity of data analysis. Artificial intelligence offers new options to manage this challenge in clinical settings. Here, we investigated whether multiclass machine learning (ML) algorithms applied to a high-dimensional panel of radiomic features from advanced MRI (advMRI) and physiological MRI (phyMRI; thus, radiophysiomics) could reliably classify contrast-enhancing brain tumors. The recently developed phyMRI technique enables the quantitative assessment of microvascular architecture, neovascularization, oxygen metabolism, and tissue hypoxia. A training cohort of 167 patients suffering from one of the five most common brain tumor entities (glioblastoma, anaplastic glioma, meningioma, primary CNS lymphoma, or brain metastasis), combined with nine common ML algorithms, was used to develop overall 135 classifiers. Multiclass classification performance was investigated using tenfold cross-validation and an independent test cohort. Adaptive boosting and random forest in combination with advMRI and phyMRI data were superior to human reading in accuracy (0.875 vs. 0.850), precision (0.862 vs. 0.798), F-score (0.774 vs. 0.740), AUROC (0.886 vs. 0.813), and classification error (5 vs. 6). The radiologists, however, showed a higher sensitivity (0.767 vs. 0.750) and specificity (0.925 vs. 0.902). We demonstrated that ML-based radiophysiomics could be helpful in the clinical routine diagnosis of contrast-enhancing brain tumors; however, a high expenditure of time and work for data preprocessing requires the inclusion of deep neural networks.

Differentiation of Glioblastoma and Brain Metastases by MRI-Based Oxygen Metabolomic Radiomics and Deep Learning.

Glioblastoma (GB) and brain metastasis (BM) are the most frequent types of brain tumors in adults. Their therapeutic management is quite different and a quick and reliable initial characterization has a significant impact on clinical outcomes. However, the differentiation of GB and BM remains a major challenge in today's clinical neurooncology due to their very similar appearance in conventional magnetic resonance imaging (MRI). Novel metabolic neuroimaging has proven useful for improving diagnostic performance but requires artificial intelligence for implementation in clinical routines. Here; we investigated whether the combination of radiomic features from MR-based oxygen metabolism ("oxygen metabolic radiomics") and deep convolutional neural networks (CNNs) can support reliably pre-therapeutic differentiation of GB and BM in a clinical setting. A self-developed one-dimensional CNN combined with radiomic features from the cerebral metabolic rate of oxygen (CMRO2) was clearly superior to human reading in all parameters for classification performance. The radiomic features for tissue oxygen saturation (mitoPO2; i.e., tissue hypoxia) also showed better diagnostic performance compared to the radiologists. Interestingly, both the mean and median values for quantitative CMRO2 and mitoPO2 values did not differ significantly between GB and BM. This demonstrates that the combination of radiomic features and DL algorithms is more efficient for class differentiation than the comparison of mean or median values. Oxygen metabolic radiomics and deep neural networks provide insights into brain tumor phenotype that may have important diagnostic implications and helpful in clinical routine diagnosis.