Publications by authors named "Walter Stummer"

Background: High-grade glioma (HGG) patients experience enormous disease burden both from tumor- and treatment-related symptoms. Exercise can improve physical fitness and quality of life (QoL); yet experience in neuro-oncology, especially with high-intensity exercise, remains limited. This study evaluated feasibility, safety, and efficacy of the intensive, structured 16-week strength and endurance program, "Active in Neuro-Oncology" (ActiNO) for HGG patients undergoing chemotherapy.

View Article and Find Full Text PDF

5-aminolevulinic acid (5-ALA) is a widely recognized and effective tool for improving tumor resections during surgical interventions but may directly interact with cells in the tumor microenvironment. Nevertheless, there remains an ongoing debate regarding the impact of 5-ALA on neural stem cells (NSCs). This study aims to investigate the effects of 5-ALA on both NSCs and oligodendrocyte progenitor cells (OPCs).

View Article and Find Full Text PDF

A 52-year-old man presented with progressive gait disturbances. MRI demonstrated a large extra-axial tumor in the cerebellopontine angle compressing the brainstem and resulting in hydrocephalus. While taking the patient's history, episodes of uncontrollable crying with sudden onset occurred multiple times.

View Article and Find Full Text PDF

Background: 5‑Aminolevulinic acid (5-ALA) is a keto-carbon amino acid frequently used in glioma surgery for fluorescence-guided resection. Additionally, cytotoxic properties of 5‑ALA can be induced via stimulation with laser light in photodynamic therapy (PDT). Preclinical in vitro and in vivo trials have also demonstrated this effect to be inducible by photon irradiation as used in radiation treatment.

View Article and Find Full Text PDF

Background: Glioblastoma (GBM) remains incurable despite multimodal therapeutic approaches. Here, we assessed the relevance of ABO blood groups for progression-free survival (PFS), overall survival (OS), and long-term survival in a large cohort of isocitrate dehydrogenase (IDH)-wildtype (wt) GBM patients.

Methods: Consecutive GBM patients (2009-2020) at a large tertiary brain tumor center were included, and clinical data were retrospectively abstracted.

View Article and Find Full Text PDF

Background: Limited amino acid availability for positron emission tomography (PET) imaging hinders therapeutic decision-making for gliomas without typical high-grade imaging features. To address this gap, we evaluated a generative artificial intelligence (AI) approach for creating synthetic O-(2-F-fluoroethyl)-l-tyrosine ([F]FET)-PET and predicting high [F]FET uptake from magnetic resonance imaging (MRI).

Methods: We trained a deep learning (DL)-based model to segment tumors in MRI, extracted radiomic features using the Python PyRadiomics package, and utilized  a Random Forest classifier to predict high [F]FET uptake.

View Article and Find Full Text PDF

Introduction: Prior research has identified temporal muscle thickness (TMT) as a prognostic marker in glioblastoma. Nonetheless, implementation in daily clinical practice is complicated due to the heterogeneity of previous studies. We performed a multicentric analysis aiming to validate recently proposed sex-specific cutoff values using a homogeneous cohort of newly diagnosed MGMT promoter methylated glioblastoma patients; we included a balanced control cohort for comparison.

View Article and Find Full Text PDF

Objective: Filter specifications for visualizing 5-aminolevulinic acid (5-ALA) tumor fluorescence are incorporated in neurosurgical wide-field microscopes. Novel exoscopes offer modified visualization technologies that should be comparable to older systems to prevent over- or underresection. In this technical note, the authors compare the fluorescence visualization technologies of three exoscopes.

View Article and Find Full Text PDF

: The utilization of non-metallic pedicle screws and rods has become a favored approach in the management of spinal tumors. An abundance of metal artifacts improves postoperative imaging and allows for precise radiation treatment planning. Under certain conditions, a vertebral body replacement (VBR) is necessary in addition to dorsal fixation.

View Article and Find Full Text PDF

Purpose: Machine Learning (ML) has become an essential tool for analyzing biomedical data, facilitating the prediction of treatment outcomes and patient survival. However, the effectiveness of ML models heavily relies on both the choice of algorithms and the quality of the input data. In this study, we aimed to develop a novel predictive model to estimate individual survival for patients diagnosed with glioblastoma (GBM), focusing on key variables such as O6-Methylguanine-DNA Methyltransferase (MGMT) methylation status, age, and sex.

View Article and Find Full Text PDF

Background: The discovery of cellular tumor networks in glioblastoma, with routes of malignant communication extending far beyond the detectable tumor margins, has highlighted the potential of supramarginal resection strategies. Retrospective data suggest that these approaches may improve long-term disease control. However, their application is limited by the proximity of critical brain regions and vasculature, posing challenges for validation in randomized trials.

View Article and Find Full Text PDF

Purpose: Sonodynamic therapy, which combines a tumor cell-selective sonosensitizer with ultrasound, is gaining attention as a promising new treatment approach for glioblastoma. The objective of this case study is to report on the first applications of 5-aminolevulinic acid (5-ALA) in combination with low-intensity, non-targeted ultrasound as neo-adjuvant treatment in therapy naïve glioblastoma.

Methods: Three patients with therapy naïve newly diagnosed glioblastoma were treated once before cytoreductive surgery with 5-ALA in combination with hemispheric, low-intensity, non-targeted ultrasound, assuming cell death to be triggered by non-ablative activation of 5-ALA-induced, tumor selective porphyrins.

View Article and Find Full Text PDF

Mutations occurring in the MeCp2, CDKL5 and BDNF genes have been linked to epileptogenesis in various epilepsy syndromes. This study employed bioinformatics analysis of transcriptomic data to examine the interrelationship among these genes in both epileptic and healthy individuals. Moreover, we assessed the expression of MeCp2, CDKL5 and BDNF at both mRNA and protein levels in human hippocampal tissues obtained from 22 patients undergoing epilepsy surgery for mesial temporal lobe epilepsy (MTLE) as well as from 25 autopsied specimens.

View Article and Find Full Text PDF

Purpose: While glioblastoma is the most common malignant brain tumor in adults, extracerebral manifestations are very rare in this highly aggressive disease with poor prognosis.

Methods: We conducted a systematic literature review in the PubMed database and complemented the data by inclusion of a case treated in our clinic. In this context, we report on a 60-year-old woman with a right frontal glioblastoma, IDH wildtype, MGMT methylated.

View Article and Find Full Text PDF
Article Synopsis
  • * Two deep learning models were developed to better correct for these issues; one uses labeled data while the other is semi-supervised, both trained on known concentrations of protoporphyrin IX (PpIX).
  • * Evaluations showed that these models had significantly higher correlation coefficients for PpIX concentration detection compared to classical methods, with the semi-supervised model also performing better on human data, reducing false positives by 36%.
View Article and Find Full Text PDF
Article Synopsis
  • This study investigates the use of Large Language Models (LLMs) to enhance clinical decision-making (CDM) skills through simulated patient-doctor interactions.
  • Medical students participated in two groups: one received only simulated conversations with AI patients, while the other also received AI-generated feedback on their performance.
  • Results showed that the feedback group significantly improved their CDM performance after a few training sessions, particularly in creating context and securing information during patient interactions.
View Article and Find Full Text PDF
Article Synopsis
  • The study aimed to determine if clinical features can predict the success of meningioma surgery by differentiating between gross total resections (GTR) and subtotal resections (STR).
  • The researchers analyzed 23 clinical features in a group of 157 patients, comparing two methods: Simpson grading and postoperative operative tumor volume (POTV) for predicting surgical outcomes.
  • Their final decision tree model demonstrated strong predictive accuracy (mean AUC of 0.885), which can assist in surgical planning and determining the need for further treatment after surgery.
View Article and Find Full Text PDF

Purpose: Lower-grade gliomas typically exhibit 5-aminolevulinic acid (5-ALA)-induced fluorescence in only 20-30% of cases, a rate that can be increased by doubling the administered dose of 5-ALA. Fluorescence can depict anaplastic foci, which can be precisely sampled to avoid undergrading. We aimed to analyze whether a deep learning model could predict intraoperative fluorescence based on preoperative magnetic resonance imaging (MRI).

View Article and Find Full Text PDF

Background: Pituitary neuroendocrine tumors (PitNET) are among the most common intracranial tumors. Despite a frequent benign course, aggressive behavior can occur. Tumor behavior is known to be under the influence of the tumor microenvironment (TME).

View Article and Find Full Text PDF

Malignant glioma resection is often the first line of treatment in neuro-oncology. During glioma surgery, the discrimination of tumor's edges can be challenging at the infiltration zone, even by using surgical adjuncts such as fluorescence guidance (e.g.

View Article and Find Full Text PDF
Article Synopsis
  • Computational neurosurgery combines artificial intelligence and computational modeling to enhance the diagnosis and treatment of neurosurgical conditions, aiming to advance clinical neurosciences.
  • The field seeks to integrate ethical considerations to ensure that the use of AI is conducted responsibly and prioritizes patient care, ultimately aiming to prevent errors in treatment.
  • This initiative serves as a guide for practitioners, ethicists, and scientists in the application of ethical standards within computational neurosurgery.
View Article and Find Full Text PDF

Background: Glioblastoma is the most common malignant brain tumor in adults. Even after maximal safe resection and adjuvant chemoradiotherapy, patients normally relapse after a few years or even months. Standard treatment for recurrent glioblastoma is not yet defined, with re-resection, re-irradiation, and systemic therapy playing key roles.

View Article and Find Full Text PDF

In recent decades, there has been increasing interest in measuring the quality of care across all medical fields, including neurosurgery. This interest aims not only to optimize care but also to reduce healthcare costs. For this purpose, different quality indicators (QIs) have been developed.

View Article and Find Full Text PDF