Citations

Abstract High-grade gliomas (HGG) have significant spatial heterogeneity, complicating therapeutic approaches. While multiparametric MRI (mpMRI) enables noninvasive assessment of this heterogeneity, most predictive models lack biological interpretability when used to classify or predict outcomes. Here, we developed an unsupervised framework that identifies novel mpMRI feature patterns through dimensional reduction, linking them to clinically relevant molecular signatures. We analyzed 18 advanced mpMRI features from 73 stereotactic biopsies across 20 HGG patients. Imaging data were normalized, followed by dimensional reduction using an iterative UMAP across varying parameters. Density-based clustering was used to generate stable image-derived clusters through consensus co-clustering. RNA-seq data from matched biopsies were analyzed using DESeq2 and GSEA to identify differentially expressed genes and enriched pathways. We then constructed a network linking specific imaging features to these transcriptional signatures. Validation was performed in an independent cohort, allowing us to extend clusters across full brain mpMRI. Unsupervised clustering of mpMRI features produced nine stable image-based clusters. These clusters were not fully explained by contrast enhancement, metabolic class, or tumor purity, shown by intraclass correlation (ICC) analyses, suggesting they capture novel biological variation. Transcriptomic analysis showed each cluster was associated with distinct gene expression patterns and enriched biological pathways, including neuronal differentiation, immune chemotaxis, and chromatin remodeling. Key imaging features (sRCBV, MTT, K2)—combined in cluster-specific patterns—enabled construction of an interpretable imaging–transcriptomic network. This was validated in an independent cohort. We then applied to full brain mpMRI, mapping spatially distinct regions to biological trends and patient outcomes. Addressing a research gap in imaging HGG, we present an interpretable network that connects specific patterns of mpMRI to clinically meaningful molecular signatures. This approach advances our understanding of tumor heterogeneity and supports future translational work in precision patient care.

Abstract BACKGROUND Glioblastoma (GBM) is a molecularly and phenotypically heterogeneous tumor, with regional invasion patterns driving treatment resistance and universal recurrence. In infiltrative regions, GBM cells adopt at least two predominant invasive phenotypes: Neuronal-enriched (NEU), with synaptic and neurodevelopment programs, and Glycolytic/Plurimetabolic (GPM), with heightened immune and metabolic activity. These subtypes differ in therapeutic vulnerability and microenvironmental interactions; features not captured by conventional contrast-enhanced (T1+C) MRI due to minimal blood-brain barrier (BBB) disruption. Our group sought to develop predictive models to distinguish variability of NEU/GPM subtypes within infiltrative GBM, utilizing graph convolutional networks (GCN), advanced multiparametric MRI (mpMRI), and spatially-matched image-localized biopsies with transcriptomic profiling. METHODS We analyzed 18 mpMRI features —including conventional, diffusion-weighted, and Dynamic Susceptibility Contrast (DSC) perfusion MRI— of 101 image-localized biopsies from MRI non-enhancing infiltrative margins of 64 GBM patient tumors. Features were hierarchically clustered per modality; each cluster was represented as a node in a Neo4j graph. Edges connected nodes sharing overlapping samples. Nodes were annotated with NEU or GPM subtype proportion. GCNs were trained to classify subtype-enriched nodes using graph topology and node features. RESULTS GCNs achieved high predictive performance on held-out test data (AUC = 0.95 for NEU, 0.88 for GPM; 70:30 train-test split) with similar 5-fold cross-validation performance and generalized well to an independent cohort (AUC > 0.85, n = 45). Importantly, performance was robust without relying on T1+C imaging, enabling prediction beyond BBB-disrupted regions. GCNs outperformed traditional classifiers (logistic regression, random forest). CONCLUSION Our results demonstrate that GCN-modeling of mpMRI features enables non-invasive prediction of invasive GBM subtypes at the tumor margin. Unlike convolutional neural networks (CNN), which rely on spatial continuity, GCNs leverage relational structure between heterogeneous imaging features, making them ideal for modeling sparse, non-grid data. This may support non-invasive regional vulnerability detection, surgical targeting, and improved response assessment.

OBJECTIVE While the relationship between smoking and subarachnoid hemorrhage is well established, data regarding the probability of detecting unruptured intracranial aneurysms (UIAs) in smokers remain sparse. The aim of this systematic review and meta-analysis is to provide a comprehensive understanding of the relationship between smoking and the likelihood of identifying UIAs in healthy asymptomatic patients who underwent brain imaging for indications unrelated to UIAs. METHODS A systematic review was conducted following the PRISMA guidelines. The PubMed and Scopus databases were searched for studies published before March 2024 that reported on the presence of UIAs in healthy asymptomatic patients who had undergone brain imaging for indications unrelated to UIAs. Three independent reviewers assessed the eligibility of all retrieved studies. Risk of bias for the included observational studies was assessed using the methodological index for non-randomized studies. Data on UIA prevalence, smoking status, and patient and aneurysm characteristics were extracted. The association between smoking and the presence of UIA was studied. A prevalence proportional meta-analysis was conducted across the included studies. A meta-analysis was performed with a random-effects model by using summary statistics from the included studies. RESULTS Six studies involving 47,788 patients who had undergone brain imaging were identified for quantitative analysis. The pooled UIA prevalence in the sample was 3.07% (95% CI 2.27%–4.16%). Patients who smoked had higher rates of aneurysm detection than the patients who did not smoke (pooled prevalence of 2.96%, 95% CI 2.68%–3.27% vs 2.23%, 95% CI 2.08%–2.39%). Although the relation between the two was not statistically significant (p = 0.06), smoking was associated with higher odds for UIA detection, yielding an OR of 1.34 (95% CI 1.07–1.67) with an I 2 value of 53%. CONCLUSIONS This study suggests a potentially higher risk of UIAs in patients who smoked than in those who never smoked. However, the results of this meta-analysis revealed that smoking was not statistically associated with higher UIA detection. This result could be explained by the limited number of studies published on this topic. If these findings reach statistical significance in future larger studies, it could justify revising guidelines to include cigarette smokers in intracranial aneurysm screening.

AbstractThe study objective was to compare the length of stay (LOS) and the proportion of one-night admissions before and after the implementation of an endocrine monitoring protocol following endoscopic transsphenoidal surgery (ETSS) for pituitary adenoma.Patients who underwent transsphenoidal pituitary adenoma resection between July 1, 2018, and September 9, 2022, were identified, and divided into two cohorts before and after the implementation of the monitoring protocol. The overall LOS and number of nights of admission were recorded. Readmission within 30 days after surgery was also recorded. The number of outpatient laboratory tests performed and the time to the first test were also noted.Thirty patients were identified in the preprotocol group and 60 in the postprotocol group. The average admission length in the preprotocol group was significantly longer than the average admission length in the postprotocol group (2.4 vs. 1.7 days, p = 0.004). The percentage of one-night admissions increased from 13 to 57% (p < 0.001). There were no significant differences in readmission rates between the two groups (p = 0.681). The number of laboratory encounters increased from a mean of 1.38 to 2.40 (p = 0.030), while the time to the first test decreased from a mean of 3.43 to 2.36 days (p = 0.049).Patients admitted after ETSS for pituitary adenoma had shorter hospital stay and greater proportion of one-night admission with the implementation of an endocrine monitoring protocol. The all-cause readmission rate was not statistically different between the two groups. Through aggressive outpatient laboratory monitoring, one-night admission for ETSS may be feasible.

BACKGROUND: Smoking and observed growth of intracranial aneurysms are known risk factors for rupture. The mechanism by which smoking increases this risk is not completely elucidated. Furthermore, an association between smoking and aneurysm growth has not been clearly defined in the literature. We hypothesize that smoking is associated with aneurysm growth, which, in turn, may serve as one of the mechanisms by which smoking drives rupture risk. METHODS: We report a systematic review of the literature in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 guidelines. Using the R software, we performed a meta-analysis to investigate the association between smoking and the growth of unruptured intracranial aneurysms. Studies on familial aneurysms and genetic syndromes known to increase the risk of aneurysms were excluded. RESULTS: Eighteen observational studies were included with a total of 3535 patients and 4289 aneurysms with a mean follow-up period ranging from 17 to 226 months. The mean age among the studies ranged from 38.4 to 73.9 years; 74% of patients were female. Ever-smoking status (odds ratio, 1.10 [95% CI, 0.87–1.38]) and current smoking status (odds ratio, 1.43 [95% CI, 0.84–2.43]) did not show a statistically significant association with growth of intracranial aneurysms. Patients currently smoking did not have a statistically significant association with the growth of intracranial aneurysms (odds ratio, 1.18 [95% CI, 0.72–1.93]) compared with patients without a smoking history. No significant association was found in patients who previously smoked compared with patients who never smoked (odds ratio, 1.46 [95% CI, 0.88–2.43]). CONCLUSIONS: Smoking is not clearly associated with the growth of unruptured intracranial aneurysms, despite trends being observed, there is no statistical association. The mechanism by which smoking increases rupture risk might not be growth. In patients for whom observation is recommended, the absence of growth over time in the setting of smoking history does not, therefore, imply protection from rupture.