Comprehensive multimodal deep learning survival prediction enabled by a transformer architecture: A multicenter study in glioblastoma.

Background: This research aims to improve glioblastoma survival prediction by integrating MR images, clinical, and molecular-pathologic data in a transformer-based deep learning model, addressing data heterogeneity and performance generalizability. Methods: We propose and evaluate a transformer-based nonlinear and nonproportional survival prediction model. The model employs self-supervised learning techniques to effectively encode the high-dimensional MRI input for integration with nonimaging data using cross-attention. To demonstrate model generalizability, the model is assessed with the time-dependent concordance index (Cdt) in 2 training setups using 3 independent public test sets: UPenn-GBM, UCSF-PDGM, and Rio Hortega University Hospital (RHUH)-GBM, each comprising 378, 366, and 36 cases, respectively. Results: The proposed transformer model achieved a promising performance for imaging as well as nonimaging data, effectively integrating both modalities for enhanced performance (UCSF-PDGM test-set, imaging Cdt 0.578, multimodal Cdt 0.672) while outperforming state-of-the-art late-fusion 3D-CNN-based models. Consistent performance was observed across the 3 independent multicenter test sets with Cdt values of 0.707 (UPenn-GBM, internal test set), 0.672 (UCSF-PDGM, first external test set), and 0.618 (RHUH-GBM, second external test set). The model achieved significant discrimination between patients with favorable and unfavorable survival for all 3 datasets (log-rank P 1.9 × 10-8, 9.7 × 10-3, and 1.2 × 10-2). Comparable results were obtained in the second setup using UCSF-PDGM for training/internal testing and UPenn-GBM and RHUH-GBM for external testing (Cdt 0.670, 0.638, and 0.621). Conclusions: The proposed transformer-based survival prediction model integrates complementary information from diverse input modalities, contributing to improved glioblastoma survival prediction compared to state-of-the-art methods. Consistent performance was observed across institutions supporting model generalizability.

Prospective Evaluation of CD45RA+/CCR7- Effector Memory T (TEMRA) Cell Subsets in Patients with Primary and Secondary Brain Tumors during Radiotherapy of the Brain within the Scope of the Prospective Glio-CMV-01 Clinical Trial.

Radiotherapy (RT) of the brain is a common treatment for patients with high-grade gliomas and brain metastases. It has previously been shown that reactivation of cytomegalovirus (CMV) frequently occurs during RT of the brain. This causes neurological decline, demands antiviral treatment, and is associated with a worse prognosis. CMV-specific T cells are characterized by a differentiated effector memory phenotype and CD45RA+ CCR7- effector memory T (TEMRA) cells were shown to be enriched in CMV seropositive individuals. In this study, we investigated the distribution of TEMRA cells and their subsets in the peripheral blood of healthy donors and, for the first time, prospectively within the scope of the prospective Glio-CMV-01 clinical trial of patients with high-grade glioma and brain metastases during radiation therapy as a potential predictive marker. First, we developed a multicolor flow cytometry-based assay to monitor the frequency and distribution of TEMRA cells in a longitudinal manner. The CMV serostatus and age were considered as influencing factors. We revealed that patients who had a reactivation of CMV have significantly higher amounts of CD8+ TEMRA cells. Further, the distribution of the subsets of TEMRA cells based on the expression of CD27, CD28, and CD57 is highly dependent on the CMV serostatus. We conclude that the percentage of CD8+ TEMRA cells out of all CD8+ T cells has the potential to serve as a biomarker for predicting the risk of CMV reactivation during RT of the brain. Furthermore, this study highlights the importance of taking the CMV serostatus into account when analyzing TEMRA cells and their subsets.

FSRT vs. SRS in Brain Metastases-Differences in Local Control and Radiation Necrosis-A Volumetric Study.

Background: While the role of stereotactic radiotherapy for brain metastases is increasing, evidence on the comparative efficacy and safety of fractionated stereotactic radiotherapy (FSRT) and single-session radiosurgery (SRS) is scarce. Methods: Longitudinal volumetric analysis was performed in a consecutive cohort of 120 patients and 190 brain metastases (>0.065 cm3 in volume / > ~5 mm in diameter) treated exclusively with FSRT (n = 98) and SRS (n = 92), respectively. A total of 972 tumor segmentations was used, averaging 5.1 time points per metastasis. Progression was defined using a volumetric extension of the RANO-BM criteria. Local control and radionecrosis were compared for lesions treated with FSRT and SRS, respectively. Results: Metastases treated with FSRT were significantly larger at baseline (mean, 4.66 vs. 0.40 cm3, p < 0.001). Biologically effective dose (BED) for metastases (α/ß = 12, linear-quadratic-cubic model) was significantly associated with local control, whereas BED for normal brain (α/ß = 2, linear-quadratic model) was significantly associated with radionecrosis. Median time to local progression was 22.9 months in the FSRT group compared to 14.5 months in the SRS group (p = 0.022). Overall radionecrosis rate at 12 months was 3.4% for FSRT and 14.8% for SRS (p = 0.010). Radionecrosis °IV requiring resection with histologic proof of radiation necrosis also was significantly reduced in the FSRT group (FSRT 0.0% vs. SRS 3.9%, p = 0.041). In multivariate analysis, FSRT was associated with reduced risk of progression (HR 0.47, p = 0.015) and reduced risk of radionecrosis (HR 0.18, p = 0.045). Conclusions: This volumetric study provides initial evidence that the improvements in therapeutic ratio expected for FSRT in larger brain metastases, might equally extend into the domain of smaller metastases, traditionally less considered for fractionated treatment. FSRT might constitute an important tool to further increase local control and reduce radionecrosis risk in stereotactic radiotherapy for brain metastases, that should be assessed in randomized intervention trials.

Brief report - Telomere length is a poor biomarker to predict 1-year mortality or cardiovascular comorbidity in patients with transcatheter aortic valve replacement.

BACKGROUND: Transcatheter aortic valve replacement (TAVR) is a therapeutic option for patients with aortic valve stenosis at increased surgical risk. Telomeres are an established marker for cellular senescence and have served to evaluate cardiovascular diseases including severe aortic valve stenosis. In our study, we hypothesized that telomere length may be a predictor for outcome and associated with comorbidities in patients with TAVR. METHODS AND RESULTS: We analyzed leucocyte telomere length from 155 patients who underwent TAVR and correlated the results with 1-year mortality and severe comorbidities. The cohort was subdivided into 3 groups according to telomere length. Although a trend for a positive correlation of telomere length with a lower EuroSCORE could be found, telomere length was not associated with survival, aortic valve opening area or cardiovascular comorbidities (peripheral, coronary or cerebrovascular disease). Interestingly, long telomeres were significantly correlated to a reduced left ventricular ejection fraction (LVEF). CONCLUSION: In elderly patients with severe aortic valve stenosis, leucocyte telomere length did not predict post-procedural survival. The correlation between long telomere length and reduced LVEF in these patients deserves further attention.