Publications by authors named "Bernhardt D"
Purpose: General practitioners (GPs) play a crucial role in providing interdisciplinary care for radiation oncology patients. This study aims to understand the specific needs and challenges faced by general practitioners in Germany when treating oncology patients.
Methods: A comprehensive web-based questionnaire with 24 items was disseminated to GPs in Germany via email using survio.
Metastasis-directed therapy (MDT) for oligometastatic breast cancer (≤ 5 metastases) has shown little effect in specific scenarios of randomized trials. Therefore, we aimed to assess outcomes after metastasis-directed stereotactic radiotherapy (SRT) in various clinical scenarios. We conducted an international retrospective cohort study in thirteen centers including breast cancer patients receiving SRT to any metastatic site.
View Article and Find Full Text PDFBackground: Post-Therapy-Pneumonitis (PTP) is a critical side effect of both, thoracic radio(chemo)therapy (R(C)T) and immune checkpoint inhibition (ICI). However, disease characteristics and patient-specific risk factors of PTP after combined R(C)T + ICI are less understood. Given that RT-triggered PTP is strongly dependent on the volume and dose of RT [1], driven by inflammatory mechanisms, we hypothesize that combination therapy of R(C)T with ICI influences the dose-volume-effect correlation for PTP.
View Article and Find Full Text PDFArticle Synopsis
- Randomized studies suggest that (ultra-)hypofractionation is just as effective as a traditional 5-week radiation schedule for treating postoperative breast cancer, but there's limited data on its long-term impact on secondary cancers.
- A study involving 20 breast cancer patients examined different radiation schedules and found that (ultra-)hypofractionation significantly reduced the risk of secondary malignancies like lung cancer, contralateral breast cancer, and soft tissue sarcoma.
- The results indicated that using (ultra-)hypofractionation could lower the risk of developing secondary malignancies more than conventional scheduling, with varying effects based on the radiation technique (3D-CRT vs. VMAT).
Background: Stereotactic radiosurgery (SRS) is an emerging alternative to whole-brain radiotherapy (WBRT) for treating multiple brain metastases (BM), reducing toxicity, and improving tumor control. The CYBER-SPACE trial compared SRS based on either SPACE or MPRAGE MRI sequence for avoiding or delaying WBRT in patients with 1-10 BM.
Methods: Patients with 1-10 untreated BM were randomized 1:1 to receive SRS of all lesions based on either SPACE or MPRAGE MRI sequences.
Article Synopsis
- Skin inflammation and conditions like moist epitheliolysis and edema are common acute side effects of breast radiotherapy (RT).
- The study aimed to evaluate the effectiveness of tissue-derived radiomics features compared to total breast volume (TBV) in predicting these side effects.
- The best predictive model used a LASSO classifier based on TBV, achieving an AUROC of 0.74, similar to the AUROC of 0.75 for TBV alone, with mammary tissue showing greater predictive power than glandular tissue.
Article Synopsis
- Molecular glioblastoma (molGB) is diagnosed as glioblastoma due to specific molecular markers, despite lacking typical grade 4 histological features, making it challenging to distinguish from lower-grade astrocytomas.
- Advanced imaging techniques revealed that molGB exhibited significantly higher relative cerebral blood volume (rCBV) and lower apparent diffusion coefficient (ADC) values compared to astrocytomas, indicating different tumor biology.
- The study suggests that quantitative imaging analysis could improve recognition of molGB in nonenhancing tumors, supporting the idea that molGB may represent an early stage of glioblastoma development.
Article Synopsis
- GBM WHO CNS Grade 4 is a tough challenge in oncology due to its aggressive nature and limitations of conventional imaging in detecting tumor recurrence.
- A study collected blood samples from seven GBM patients at various stages to identify gene-based biomarkers for early detection of recurrence, utilizing next-generation sequencing for analysis.
- Results showed significant variability in gene expression among patients, with a promising indication that measuring gene expressions in whole blood could be a feasible alternative to traditional methods involving exosome isolation.
Article Synopsis
- - Meningiomas are the most common primary brain tumors, with most being benign, but around 25% are higher-grade and require better risk assessment through an integrated risk score (IRS) based on tumor biology.
- - The study involved 160 patients and utilized machine learning with preoperative MRI scans to develop classifiers that predict the IRS, achieving a high accuracy of 90% when distinguishing low-risk from medium/high-risk patients.
- - The results highlight that specific imaging characteristics, like "sphericity," can effectively predict the molecular low-risk classification of meningiomas, making critical prognostic information more accessible through imaging techniques.
Article Synopsis
- Spatial intratumoral heterogeneity poses challenges for assessing glioblastoma treatment responses, but multimodal imaging and advanced analysis can help address this issue.
- A study categorized tumor voxels from 61 patients into true tumor progression or pseudoprogression using supervoxels and a Random Forest classifier, achieving 80% accuracy in identifying these categories.
- The analysis highlighted the importance of FET-PET features while also showing that other imaging factors like cerebral blood volume and T1-weighted imaging contributed significantly, guiding future personalized treatment strategies.
Article Synopsis
- Surgical resection is the primary treatment for patients with large or symptomatic brain metastases, but there's still a risk of local failure, prompting the development of a prediction tool to identify those at high risk.
- Data from the AURORA study included 253 patients for training and 99 for external testing, utilizing radiomic features from MRI scans to enhance prediction accuracy.
- The elastic net regression model combining radiomic and clinical features showed a significant improvement in predicting local failure, with lower risk groups experiencing only 9% failure at 24 months compared to 74% in high-risk groups, suggesting potential for improved patient follow-up and treatment.
Article Synopsis
- The study focuses on developing a Deep Learning-based automatic segmentation (DLBAS) algorithm to improve the process of defining volume of interest (VOI) for radiomic analyses in extremity soft tissue sarcomas, addressing issues like time consumption and variability among observers.* -
- The DLBAS was trained on 157 patients and tested on 87, comparing its automatic segmentations to manual delineations by radiation oncologists and residents, showing promising median dice similarity coefficients that indicate high accuracy in VOI predictions.* -
- Despite achieving high reproducibility for radiomics feature extraction, the clinical applicability of the DLBAS predictions for radiotherapy planning was found to be limited, with radiation oncologists deeming them suitable in only
Article Synopsis
- This study investigates the relationship between glioblastoma (GBM) tumors and key neurogenic areas in the brain (SVZ and SGZ) to understand their impact on patient survival outcomes.
- Using advanced segmentation techniques, researchers analyzed data from 177 GBM patients and found that tumor proximity to the SVZ negatively correlated with survival, establishing a critical distance of 14 mm.
- Unlike previous studies, this research did not find significant relationships between SVZ contact and factors like tumor multifocality or MGMT promotor methylation, suggesting the need for more objective assessments in future GBM studies.
Article Synopsis
- Brain metastases (BM) are increasingly common due to improved treatments and longer survival rates, with ongoing debates about the role of postoperative tumor volume in patient outcomes.
- This study analyzed patients with surgically treated single BM to see how tumor resection and remaining tumor burden affected overall survival (OS).
- Of 340 patients studied, the type of BM (solitary vs. singular) significantly influenced OS, with solitary BM patients generally having better outcomes, especially when complete tumor resection was achieved.
Article Synopsis
- * Out of 44 patients analyzed, the median overall survival (OS) after the second surgery was found to be 14 months, but Re-RT did not show a significant impact on OS. However, better OS was linked to MGMT promoter methylation status and longer intervals between treatments.
- * The conclusion emphasizes the importance of personalized treatment strategies based on MGMT status and treatment intervals for improving outcomes in rGBM patients, urging further research to clarify the
Article Synopsis
- The study focuses on improving treatment for glioblastoma, a challenging brain cancer, by validating a new computational tumor growth model for personalized therapy.
- Researchers analyzed data from 124 TCGA patients and 397 UCSF patients to find links between clinical outcomes and factors related to tumor growth and genetics.
- Results indicate that certain growth parameters are significantly linked to patient survival and that the model may enhance radiation treatment planning without increasing radiation exposure.
Article Synopsis
- MRI scans take a long time to get, especially in places with fewer resources, so researchers want to make it faster using a special computer program called a deep convolutional neural network (dCNN).
- They studied information from a lot of patients with a type of brain cancer called glioblastoma to help train the dCNN to create better MRI images quickly by using less data.
- Their tests showed that the dCNN-created images were very similar to the original ones, and it worked well when looking at important cancer details in the scans.
Objectives: Post-therapy pneumonitis (PTP) is a relevant side effect of thoracic radiotherapy and immunotherapy with checkpoint inhibitors (ICI). The influence of the combination of both, including dose fractionation schemes on PTP development is still unclear. This study aims to improve the PTP risk estimation after radio(chemo)therapy (R(C)T) for lung cancer with and without ICI by investigation of the impact of dose fractionation on machine learning (ML)-based prediction.
View Article and Find Full Text PDFBackground: A reduced Karnofsky performance score (KPS) often leads to the discontinuation of surgical and adjuvant therapy, owing to a lack of evidence of survival and quality of life benefits. This study aimed to examine the clinical and treatment outcomes of patients with KPS < 70 after neurosurgical resection and identify prognostic factors associated with better survival.
Methods: Patients with a preoperative KPS < 70 who underwent surgical resection for newly diagnosed brain metastases (BM) between 2007 and 2020 were retrospectively analyzed.
Article Synopsis
- The study compares amino acid PET scans and conventional MRI to predict overall survival in patients with recurrent high grade glioma undergoing bevacizumab therapy.
- Five studies with 72 patients revealed that PET-positive patients had significantly lower median survival compared to PET-negative patients.
- The findings suggest that AA-PET is more effective than RANO MRI in predicting overall survival, with higher sensitivity and area under the curve values.
Background: Novel radiotherapeutic modalities using carbon ions provide an increased relative biological effectiveness (RBE) compared to photons, delivering a higher biological dose while reducing radiation exposure for adjacent organs. This prospective phase 2 trial investigated bimodal radiotherapy using photons with carbon-ion (C12)-boost in patients with WHO grade 2 meningiomas following subtotal resection (Simpson grade 4 or 5).
Methods: A total of 33 patients were enrolled from July 2012 until July 2020.
Article Synopsis
- Photonic integrated circuits (PICs) are transforming information tech by offering faster and more efficient data processing and communication, but creating them at a nanoscale precision is tricky due to the need for precise adjustments.
- This research explores using automated silicon ion implantation to achieve scalable and stable photonic computational memories, allowing for fine-tuning of components with minimal loss.
- The study showcases spectrally aligned photonic memory and computing systems that enable large-scale matrix multiplication, supporting advanced integrated architectures that can work with multiple wavelengths.
Background: Despite advances in treatment for brain metastases (BMs), the prognosis for recurrent BMs remains poor and requires further research to advance clinical management and improve patient outcomes. In particular, data addressing the impact of tumor volume and surgical resection with regard to survival remain scarce.
Methods: Adult patients with recurrent BMs between December 2007 and December 2022 were analyzed.
Patients suffering from painful spinal bone metastases (PSBMs) often undergo palliative radiation therapy (RT), with an efficacy of approximately two thirds of patients. In this exploratory investigation, we assessed the effectiveness of machine learning (ML) models trained on radiomics, semantic and clinical features to estimate complete pain response. Gross tumour volumes (GTV) and clinical target volumes (CTV) of 261 PSBMs were segmented on planning computed tomography (CT) scans.
View Article and Find Full Text PDFBackground: Many automatic approaches to brain tumor segmentation employ multiple magnetic resonance imaging (MRI) sequences. The goal of this project was to compare different combinations of input sequences to determine which MRI sequences are needed for effective automated brain metastasis (BM) segmentation.
Methods: We analyzed preoperative imaging (T1-weighted sequence ± contrast-enhancement (T1/T1-CE), T2-weighted sequence (T2), and T2 fluid-attenuated inversion recovery (T2-FLAIR) sequence) from 339 patients with BMs from seven centers.