Diffuse midline glioma-H3K27M mutant (DMG) and glioblastoma (GBM) are the most lethal brain tumors that primarily occur in pediatric and adult patients, respectively. Both tumors exhibit significant heterogeneity, shaped by distinct genetic/epigenetic drivers, transcriptional programs including RNA splicing, and microenvironmental cues in glioma niches. However, the spatial organization of cellular states and niche-specific regulatory programs remain to be investigated. Here, we perform a spatial profiling of DMG and GBM combining short- and long-read spatial transcriptomics, and single-cell transcriptomic datasets. We identify clinically relevant transcriptional programs, RNA isoform diversity, and multi-cellular ecosystems across different glioma niches. We find that while the tumor core enriches for oligodendrocyte precursor-like cells, radial glial stem-like (RG-like) cells are enriched in the neuron-rich invasive niche in both DMG and GBM. Further, we identify niche-specific regulatory programs for RG-like cells, and functionally confirm that FAM20C mediates invasive growth of RG-like cells in a neuron-rich microenvironment in a human neural stem cell derived orthotopic DMG model. Together, our results provide a blueprint for understanding the spatial architecture and niche-specific vulnerabilities of DMG and GBM.

Background: H3K27M mutation status significantly affects the prognosis of patients with diffuse midline gliomas (DMGs), but this tumor presents a high risk of pathological acquisition. We aimed to construct a fully automated model for predicting the H3K27M alteration status of DMGs based on deep learning using whole‐brain MRI.Methods: DMG patients from West China Hospital of Sichuan University (WCHSU; n = 200) and Chengdu Shangjin Nanfu Hospital (CSNH; n = 35) who met the inclusion and exclusion criteria from February 2016 to April 2022 were enrolled as the training and external test sets, respectively. To adapt the model to the human head MRI scene, we use normal human head MR images to pretrain the model. The classification and tumor segmentation tasks are naturally related, so we conducted cotraining for the two tasks to enable information interaction between them and improve the accuracy of the classification task.Results: The average classification accuracies of our model on the training and external test sets was 90.5% and 85.1%, respectively. Ablation experiments showed that pretraining and cotraining could improve the prediction accuracy and generalization performance of the model. In the training and external test sets, the average areas under the receiver operating characteristic curve (AUROCs) were 94.18% and 87.64%, and the average areas under the precision‐recall curve (AUPRC) were 93.26% and 85.4%.Conclusions: The developed model achieved excellent performance in predicting the H3K27M alteration status in DMGs, and its good reproducibility and generalization were verified in the external dataset.

Background and Objective: High‐grade glioma (HGG) is known to be characterized by a high degree of malignancy and a worse prognosis. The classical treatment is safe resection supplemented by radiotherapy and chemotherapy. Tumor treating fields (TTFields), an emerging physiotherapeutic modality that targets malignant solid tumors using medium‐frequency, low‐intensity, alternating electric fields to interfere with cell division, have been used for the treatment of new diagnosis of glioblastoma, however, their administration in HGG requires further clinical evidence. The efficacy and safety of TTFields in Chinese patients with HGG were retrospectively evaluated by us in a single center.Methods: We enrolled and analyzed 52 patients with newly diagnosed HGG undergoing surgery and standard chemoradiotherapy regimens from December 2019 to June 2022, and followed them until June 2023. Based on whether they used TTFields, they were divided into a TTFields group and a non‐TTFields group. Progression‐free survival (PFS) and overall survival (OS) were compared between the two groups.Results: There were 26 cases in the TTFields group and 26 cases in the non‐TTFields group. In the TTFields group, the median PFS was 14.2 months (95% CI: 9.50–18.90), the median OS was 19.7 months (95% CI: 14.95–24.25) , the median interval from surgery to the start of treatment with TTFields was 2.47 months (95% CI: 1.47–4.13), and the median duration of treatment with TTFields was 10.6 months (95% CI: 9.57–11.63). 15 (57.69%) patients experienced an adverse event and no serious adverse event was reported. In the non‐TTFields group, the median PFS was 9.57 months (95% CI: 6.23–12.91) and the median OS was 16.07 months (95% CI: 12.90–19.24). There was a statistically significant difference in PFS (p = 0.005) and OS (p = 0.007) between the two groups.Conclusions: In this retrospective analysis, TTFields were observed to improve newly diagnosed HGG patients' median PFS and OS. Compliance was much higher than reported in clinical trials and safety remained good.

Background: The prognosis of diffuse midline glioma (DMG) patients with H3K27M (H3K27M-DMG) alterations is poor; however, a model that encourages accurate prediction of prognosis for such lesions on an individual basis remains elusive. We aimed to construct an H3K27M-DMG survival model based on DeepSurv to predict patient prognosis. Methods: Patients recruited from a single center were used for model training, and patients recruited from another center were used for external validation. Univariate and multivariate Cox regression analyses were used to select features. Four machine learning models were constructed, and the consistency index (C-index) and integrated Brier score (IBS) were calculated. We used the receiver operating characteristic curve (ROC) and area under the receiver operating characteristic (AUC) curve to assess the accuracy of predicting 6-month, 12-month, 18-month and 24-month survival rates. A heatmap of feature importance was used to explain the results of the four models. Results: We recruited 113 patients in the training set and 23 patients in the test set. We included tumor size, tumor location, Karnofsky Performance Scale (KPS) score, enhancement, radiotherapy, and chemotherapy for model training. The accuracy of DeepSurv prediction is highest among the four models, with C-indexes of 0.862 and 0.811 in the training and external test sets, respectively. The DeepSurv model had the highest AUC values at 6 months, 12 months, 18 months and 24 months, which were 0.970 (0.919–1), 0.950 (0.877–1), 0.939 (0.845–1), and 0.875 (0.690–1), respectively. We designed an interactive interface to more intuitively display the survival probability prediction results provided by the DeepSurv model. Conclusion: The DeepSurv model outperforms traditional machine learning models in terms of prediction accuracy and robustness, and it can also provide personalized treatment recommendations for patients. The DeepSurv model may provide decision-making assistance for patients in formulating treatment plans in the future.

Objective : Blood-blister aneurysms (BBAs) of the internal carotid artery (ICA) are challenging lesions with high morbidity and mortality rates. Although research on BBAs is well documented in different populations, the study of BBAs in the Tibetan population is extremely rare. This study aimed to evaluate the characteristics of BBAs and analyze the treatment modalities and long-term outcomes in the Tibetan population in comparison with the Han population.Methods : The characteristics of patients with BBAs of the ICA from January 2009 to January 2021 at our institution were reviewed. The features of aneurysms, treatment modalities, complications, and follow-up outcomes were retrospectively analyzed.Results : A total of 130 patients (41 Tibetan and 89 Han patients) with BBAs of the ICA who underwent treatment were enrolled. Compared with the Han group, the Tibetan group significantly demonstrated a high ratio of BBAs among ICAs (8.6%, 41/477 vs. 1.6%, 89/5563; p<0.05), a high ratio of vasospasm (34.1%, 14/41 vs. 6.7%, 6/89; p=0.001), a high risk of ischemic events (43.9%, 18/41 vs. 22.5%, 20/89; p<0.05), and a low ratio of good outcomes (modified Rankin scale, 0–2) at the 1-year follow-up (51.2%, 21/41 vs. 74.2%, 66/89; p<0.05). The multivariate regression model showed that ischemic events significantly contributed to the prediction of outcomes at 1 year. Further analysis revealed that microsurgery and vasospasm were associated with ischemic events.Conclusion : In comparison with Han patients, the Tibetan population had a high ratio of BBA occurrence, a high incidence of ischemic events, and a high ratio of poor outcomes. The endovascular approach showed more benefits in BBA patients.