Postoperative Injection of a Triptolide-Preloaded Hydrogel Prevents the Recurrence of Glioblastoma by Dual-Pathway Activation of Ferroptosis.

Glioblastoma (GBM) recurrence leads to high mortality, which remains a major concern in clinical therapy. Herein, an injectable triptolide (TP)-preloaded hydrogel (TP@DNH) accompanied by a postoperative injection strategy is developed to prevent the recurrence of GBM. With a potential inhibitor of the NRF2/SLC7A11/GPX4 axis, it is demonstrated that TP can deactivate glutathione peroxidase 4 (GPX4) from the source of glutathione (GSH) biosynthesis, thereby activating ferroptosis in GBM cells by blocking the neutralization of intracellular lipid peroxide (LPO). Based on acid-sensitive Fe3+/tannic acid (TA) metal-phenolic networks (MPNs), the TP@DNH hydrogel can induce the effective generation of reactive oxygen species (ROS) through Fe3+/TA-mediated Fenton reaction and achieve controllable release of TP in resected GBM cavity. Due to ROS generation and GPX4 deactivation, postoperative injection of TP@DNH can achieve high-level ferroptosis through dual-pathway LPO accumulation, remarkably suppressing the growth of recurrent GBM and prolonging the overall survival in orthotopic GBM relapse mouse model. This work provides an alternative paradigm for regulating ferroptosis in the postoperative treatment of GBM.

Tuning the fluidity and protein corona of ultrasound-responsive liposomal nanovaccines to program T cell immunity in mice.

Inducing high levels of antigen-specific CD8α+ T cells in the tumor is beneficial for cancer immunotherapy, but achieving this in a safe and effective manner remains challenging. Here, we have developed a designer liposomal nanovaccine containing a sonosensitizer (LNVS) to efficiently program T cell immunity in mice. Following intravenous injection, LNVS accumulates in the spleen in a protein corona and fluidity-dependent manner, leading to greater frequencies of antigen-specific CD8α+ T cells than soluble vaccines (the mixture of antigens and adjuvants). Meanwhile, some LNVS passively accumulates in the tumor, where it responds to ultrasound (US) to increase the levels of chemokines and adhesion molecules that are beneficial for recruiting CD8α+ T cells to the tumor. LNVS + US induces higher levels of intratumoral antitumor T cells than traditional sonodynamic therapy, regresses established mouse MC38 tumors and orthotopic cervical cancer, and protects cured mice from relapse. Our platform sheds light on the importance of tuning the fluidity and protein corona of naovaccines to program T cell immunity in mice and may inspire new strategies for cancer immunotherapy.

Low serum CTRP3 is related to more severe distal symmetric polyneuropathy in type 2 diabetes patients.

BACKGROUND: Distal symmetric polyneuropathy (DSPN) is one of the most common chronic complications in patients with type 2 diabetes mellitus (T2DM). Our previous study found that serum C1q tumor necrosis factor-related protein 3 (CTRP3) levels were decreased in type 2 diabetic patients. Thus, this study was designed to reveal the relationship between low serum CTRP3 and the prevalence and severity of DSPN. METHODS: A total of 178 cases of patients with T2DM were enrolled in the study. The subjects were divided into the DSPN group (n = 89) and the non-DSPN group (n = 89). Both anthropometric parameters and neurologic symptoms were recorded. Furthermore, neurologic signs, the neuropathy symptom score (NSS), and the neuropathy disability score (NDS) were assessed. Biochemical indexes, fasting insulin, and C peptide were measured. Serum CTRP3 concentrations were assayed using the ELISA method. RESULTS: Serum CTRP3 levels decreased significantly in the DSPN group compared with the non-DSPN group (P < 0.05). CTRP3 was negatively associated with the number of positive signs, NSS score, and NDS score in patients with DSPN (all P < 0.05). Interestingly, the higher the NSS score or NDS score, the lower were the levels of serum CTRP3 (all P < 0.05). Moreover, patients with lower CTRP3 levels (< 7.58ng/ml) had a higher rate of neurologic signs (all P < 0.05). Binary logistic regression analysis showed that CTRP3 independently predicted the occurrence of DSPN (ß = -0.316, P < 0.001). ROC curve analysis revealed that the best cut-off value of CTRP3 for the prediction of DSPN was 7.55ng/ml (sensitivity 78.7%, specificity 79.8%), the area under the curve (AUC) was 0.763 (95% CI 0.689-0.838, P < 0.001). CONCLUSION: Low serum CTRP3 could be a predictor for the occurrence and progression of DSPN in Chinese patients with T2DM.

Multi-omics and pharmacological characterization of patient-derived glioma cell lines.

Glioblastoma (GBM) is the most common brain tumor and remains incurable. Primary GBM cultures are widely used tools for drug screening, but there is a lack of genomic and pharmacological characterization for these primary GBM cultures. Here, we collect 50 patient-derived glioma cell (PDGC) lines and characterize them by whole genome sequencing, RNA sequencing, and drug response screening. We identify three molecular subtypes among PDGCs: mesenchymal (MES), proneural (PN), and oxidative phosphorylation (OXPHOS). Drug response profiling reveals that PN subtype PDGCs are sensitive to tyrosine kinase inhibitors, whereas OXPHOS subtype PDGCs are sensitive to histone deacetylase inhibitors, oxidative phosphorylation inhibitors, and HMG-CoA reductase inhibitors. PN and OXPHOS subtype PDGCs stably form tumors in vivo upon intracranial transplantation into immunodeficient mice, whereas most MES subtype PDGCs fail to form tumors in vivo. In addition, PDGCs cultured by serum-free medium, especially long-passage PDGCs, carry MYC/MYCN amplification, which is rare in GBM patients. Our study provides a valuable resource for understanding primary glioma cell cultures and clinical translation and highlights the problems of serum-free PDGC culture systems that cannot be ignored.

Rapid Glutathione Analysis with SERS Microneedles for Deep Glioblastoma Tissue Differentiation.

Rapid tissue differentiation at the molecular level is a prerequisite for precise surgical resection, which is of special value for the treatment of malignant tumors, such as glioblastoma (GBM). Herein, a SERS-active microneedle is prepared by modifying glutathione (GSH)-responsive molecules, 5,5'-dithiobis(2-nitrobenzoic acid) (DTNB), on the surface of Au@Ag substrates for the distinction of different GBM tissues. Since the Raman signals on the surface of the DTNB@Au@Ag microneedle can be collected by both portable and benchtop Raman spectrometers, the distribution of GSH in different tissues at centimeter scale can be displayed through Raman spectroscopy and Raman imaging, and the entire analysis process can be accomplished within 12 min. Accordingly, in vivo brain tissues of orthotopic GBM xenograft mice and ex vivo tissues of GBM patients are accurately differentiated with the microneedle, and the results are well consistent with tissue staining and postoperative pathological reports. In addition, the outline of tumor, peritumoral, and normal tissues can be indicated by the DTNB@Au@Ag microneedle for at least 56 days. Considering that the tumor tissues are quickly discriminated at the molecular level without the restriction of depth, the DTNB@Au@Ag microneedle is promising to be a powerful intraoperative diagnostic tool for surgery navigation.

Radiation-primed TGF-ß trapping by engineered extracellular vesicles for targeted glioblastoma therapy.

The poor outcome of glioblastoma multiforme (GBM) treated with immunotherapy is attributed to the profound immunosuppressive tumor microenvironment (TME) and the lack of effective delivery across the blood-brain barrier. Radiation therapy (RT) induces an immunogenic antitumor response that is counteracted by evasive mechanisms, among which transforming growth factor-ß (TGF-ß) activation is the most prominent factor. We report an extracellular vesicle (EV)-based nanotherapeutic that traps TGF-ß by expressing the extracellular domain of the TGF-ß type II receptor and targets GBM by decorating the EV surface with RGD peptide. We show that short-burst radiation dramatically enhanced the targeting efficiency of RGD peptide-conjugated EVs to GBM, while the displayed TGF-ß trap reversed radiation-stimulated TGF-ß activation in the TME, offering a synergistic effect in the murine GBM model. The combined therapy significantly increased CD8+ cytotoxic T cells infiltration and M1/M2 macrophage ratio, resulting in the regression of tumor growth and prolongation of overall survival. These results provide an EV-based therapeutic strategy for immune remodeling of the GBM TME and eradication of therapy-resistant tumors, further supporting its clinical translation.

A Virus-Inspired Inhalable Liponanogel Induces Potent Antitumor Immunity and Regression in Metastatic Lung Tumors.

Pulmonary delivery of immunostimulatory agents such as poly(I:C) to activate double-stranded RNA sensors MDA5 and RIG-I within lung-resident antigen-presenting cells is a potential strategy to enhance antitumor immunity by promoting type I interferon secretion. Nevertheless, following pulmonary delivery, poly(I:C) suffers from rapid degradation and poor endosomal escape, thus limiting its potency. Inspired by the structure of a virus that utilizes internal viral proteins to tune the loading and cytosolic delivery of viral nucleic acids, we developed a liponanogel (LNG)-based platform to overcome the delivery challenges of poly(I:C). The LNG comprised an anionic polymer hyaluronic acid-based nanogel core coated by a lipid shell, which served as a protective layer to stabilize the nanogel core in the lungs. The nanogel core was protonated within acidic endosomes to enhance the endosomal membrane permeability and cytosolic delivery of poly(I:C). After pulmonary delivery, LNG-poly(I:C) induced 13.7-fold more IFNß than poly(I:C) alone and two-fold more than poly(I:C) loaded in the state-of-art lipid nanoparticles [LNP-poly(I:C)]. Additionally, LNG-poly(I:C) induced more potent CD8+ T-cell immunity and stronger therapeutic effects than LNP-poly(I:C). The combination of LNG-poly(I:C) and PD-L1 targeting led to regression of established lung metastases. Due to the ease of manufacturing and the high biocompatibility of LNG, pulmonary delivery of LNG may be broadly applicable to the treatment of different lung tumors and may spur the development of innovative strategies for cancer immunotherapy. Significance: Pulmonary delivery of poly(I:C) with a virus-inspired inhalable liponanogel strongly activates cytosolic MDA5 and RIG-I and stimulates antitumor immunity, representing a promising strategy for safe and effective treatment of metastatic lung tumors.

Modeling and experimental study of the intervention forces between the guidewire and blood vessels.

A profound investigation of the interaction mechanics between blood vessels and guidewires is necessary to achieve safe intervention. An interactive force model between guidewires and blood vessels is established based on cardiovascular fluid dynamics theory and contact mechanics, considering two intervention phases (straight intervention and contact intervention at a corner named "J-vessel"). The contributing factors of the force model, including intervention conditions, guidewire characteristics, and intravascular environment, are analyzed. A series of experiments were performed to validate the availability of the interactive force model and explore the effects of influential factors on intervention force. The intervention force data were collected using a 2-DOF mechanical testing system instrumented with a force sensor. The guidewire diameter and material were found to significantly impact the intervention force. Additionally, the intervention force was influenced by factors such as blood viscosity, blood vessel wall thickness, blood flow velocity, as well as the interventional velocity and interventional mode. The experiment of the intervention in a coronary artery physical vascular model confirms the practicality validation of the predicted force model and can provide an optimized interventional strategy for vascular interventional surgery. The enhanced intervention strategy has resulted in a considerable reduction of approximately 21.97 % in the force exerted on blood vessels, effectively minimizing the potential for complications associated with the interventional surgery.

IFI35 regulates non-canonical NF-κB signaling to maintain glioblastoma stem cells and recruit tumor-associated macrophages.

Glioblastoma (GBM) is the most aggressive malignant primary brain tumor characterized by a highly heterogeneous and immunosuppressive tumor microenvironment (TME). The symbiotic interactions between glioblastoma stem cells (GSCs) and tumor-associated macrophages (TAM) in the TME are critical for tumor progression. Here, we identified that IFI35, a transcriptional regulatory factor, plays both cell-intrinsic and cell-extrinsic roles in maintaining GSCs and the immunosuppressive TME. IFI35 induced non-canonical NF-kB signaling through proteasomal processing of p105 to the DNA-binding transcription factor p50, which heterodimerizes with RELB (RELB/p50), and activated cell chemotaxis in a cell-autonomous manner. Further, IFI35 induced recruitment and maintenance of M2-like TAMs in TME in a paracrine manner. Targeting IFI35 effectively suppressed in vivo tumor growth and prolonged survival of orthotopic xenograft-bearing mice. Collectively, these findings reveal the tumor-promoting functions of IFI35 and suggest that targeting IFI35 or its downstream effectors may provide effective approaches to improve GBM treatment.

Design and development of a personalized virtual reality-based training system for vascular intervention surgery.

BACKGROUND AND OBJECTIVES: Virtual training has emerged as an exceptionally effective approach for training healthcare practitioners in the field of vascular intervention surgery. By providing a simulated environment and blood vessel model that enables repeated practice, virtual training facilitates the acquisition of surgical skills in a safe and efficient manner for trainees. However, the current state of research in this area is characterized by limitations in the fidelity of blood vessel and guidewire models, which restricts the effectiveness of training. Additionally, existing approaches lack the necessary real-time responsiveness and precision, while the blood vessel models suffer from incompleteness and a lack of scientific rigor. METHODS: To address these challenges, this paper integrates position-based dynamics (PBD) and its extensions, shape matching, and Cosserat elastic rods. By combining these approaches within a unified particle framework, accurate and realistic deformation simulation of personalized blood vessel and guidewire models is achieved, thereby enhancing the training experience. Furthermore, a multi-level progressive continuous collision detection method, leveraging spatial hashing, is proposed to improve the accuracy and efficiency of collision detection. RESULTS: Our proposed blood vessel model demonstrated acceptable performance with the reduced deformation simulation response times of 7 ms, improving the real-time capability at least of 43.75 %. Experimental validation confirmed that the guidewire model proposed in this paper can dynamically adjust the density of its elastic rods to alter the degree of bending and torsion. It also exhibited a deformation process comparable to that of real guidewires, with an average response time of 6 ms. In the interaction of blood vessel and guidewire models, the simulator blood vessel model used for coronary vascular intervention training exhibited an average response time of 15.42 ms, with a frame rate of approximately 64 FPS. CONCLUSIONS: The method presented in this paper achieves deformation simulation of both vascular and guidewire models, demonstrating sufficient real-time performance and accuracy. The interaction efficiency between vascular and guidewire models is enhanced through the unified simulation framework and collision detection. Furthermore, it can be integrated with virtual training scenarios within the system, making it suitable for developing more advanced vascular interventional surgery training systems.

Machine-Learning and Radiomics-Based Preoperative Prediction of Ki-67 Expression in Glioma Using MRI Data.

BACKGROUND: Gliomas are the most common primary brain tumours and constitute approximately half of all malignant glioblastomas. Unfortunately, patients diagnosed with malignant glioblastomas typically survive for less than a year. In light of this circumstance, genotyping is an effective means of categorising gliomas. The Ki67 proliferation index, a widely used marker of cellular proliferation in clinical contexts, has demonstrated potential for predicting tumour classification and prognosis. In particular, magnetic resonance imaging (MRI) plays a vital role in the diagnosis of brain tumours. Using MRI to extract glioma-related features and construct a machine learning model offers a viable avenue to classify and predict the level of Ki67 expression. METHODS: This study retrospectively collected MRI data and postoperative immunohistochemical results from 613 glioma patients from the First Affliated Hospital of Nanjing Medical University. Subsequently, we performed registration and skull stripping on the four MRI modalities: T1-weighted (T1), T2-weighted (T2), T1-weighted with contrast enhancement (T1CE), and Fluid Attenuated Inversion Recovery (FLAIR). Each modality's segmentation yielded three distinct tumour regions. Following segmentation, a comprehensive set of features encompassing texture, first-order, and shape attributes were extracted from these delineated regions. Feature selection was conducted using the least absolute shrinkage and selection operator (LASSO) algorithm with subsequent sorting to identify the most important features. These selected features were further analysed using correlation analysis to finalise the selection for machine learning model development. Eight models: logistic regression (LR), naive bayes, decision tree, gradient boosting tree, and support vector classification (SVM), random forest (RF), XGBoost, and LightGBM were used to objectively classify Ki67 expression. RESULTS: In total, 613 patients were enroled in the study, and 24,455 radiomic features were extracted from each patient's MRI. These features were eventually reduced to 36 after LASSO screening, RF importance ranking, and correlation analysis. Among all the tested machine learning models, LR and linear SVM exhibited superior performance. LR achieved the highest area under the curve score of 0.912 ± 0.036, while linear SVM obtained the top accuracy with a score of 0.884 ± 0.031. CONCLUSION: This study introduced a novel approach for classifying Ki67 expression levels using MRI, which has been proven to be highly effective. With the LR model at its core, our method demonstrated its potential in signalling a promising avenue for future research. This innovative approach of predicting Ki67 expression based on MRI features not only enhances our understanding of cell activity but also represents a significant leap forward in brain glioma research. This underscores the potential of integrating machine learning with medical imaging to aid in the diagnosis and prognosis of complex diseases.

Reactivating PTEN to impair glioma stem cells by inhibiting cytosolic iron-sulfur assembly.

Glioblastoma, the most lethal primary brain tumor, harbors glioma stem cells (GSCs) that not only initiate and maintain malignant phenotypes but also enhance therapeutic resistance. Although frequently mutated in glioblastomas, the function and regulation of PTEN in PTEN-intact GSCs are unknown. Here, we found that PTEN directly interacted with MMS19 and competitively disrupted MMS19-based cytosolic iron-sulfur (Fe-S) cluster assembly (CIA) machinery in differentiated glioma cells. PTEN was specifically succinated at cysteine (C) 211 in GSCs compared with matched differentiated glioma cells. Isotope tracing coupled with mass spectrometry analysis confirmed that fumarate, generated by adenylosuccinate lyase (ADSL) in the de novo purine synthesis pathway that is highly activated in GSCs, promoted PTEN C211 succination. This modification abrogated the interaction between PTEN and MMS19, reactivating the CIA machinery pathway in GSCs. Functionally, inhibiting PTEN C211 succination by reexpressing a PTEN C211S mutant, depleting ADSL by shRNAs, or consuming fumarate by the US Food and Drug Administration-approved prescription drug N-acetylcysteine (NAC) impaired GSC maintenance. Reexpressing PTEN C211S or treating with NAC sensitized GSC-derived brain tumors to temozolomide and irradiation, the standard-of-care treatments for patients with glioblastoma, by slowing CIA machinery-mediated DNA damage repair. These findings reveal an immediately practicable strategy to target GSCs to treat glioblastoma by combination therapy with repurposed NAC.

RUNX1/NPM1/H3K4me3 complex contributes to extracellular matrix remodeling via enhancing FOSL2 transcriptional activation in glioblastoma.

Extracellular matrix (ECM) remodeling has been implicated in the tumor malignant progression and immune escape in glioblastoma (GBM). Runt-related transcription factor 1 (RUNX1) is a vital transcriptional factor for promoting tumorigenesis and invasion in mesenchymal subtype of GBM. But the correlation between RUNX1 and ECM genes expression and regulatory mechanism of RUNX1 on ECM genes expression remain poorly understood to date. In this study, by using integral analysis of chromatin immunoprecipitation-sequencing and RNA sequencing, we reported that RUNX1 positively regulated the expression of various ECM-related genes, including Fibronectin 1 (FN1), Collagen type IV alpha 1 chain (COL4A1), and Lumican (LUM), in GBM. Mechanistically, we demonstrated that RUNX1 interacted with Nucleophosmin 1 (NPM1) to maintain the chromatin accessibility and facilitate FOS Like 2, AP-1 Transcription Factor Subunit (FOSL2)-mediated transcriptional activation of ECM-related genes, which was independent of RUNX1's transcriptional function. ECM remodeling driven by RUNX1 promoted immunosuppressive microenvironment in GBM. In conclusion, this study provides a novel mechanism of RUNX1 binding to NPM1 in driving the ECM remodeling and GBM progression.

The Chalcidoidea bush of life: evolutionary history of a massive radiation of minute wasps.

Chalcidoidea are mostly parasitoid wasps that include as many as 500 000 estimated species. Capturing phylogenetic signal from such a massive radiation can be daunting. Chalcidoidea is an excellent example of a hyperdiverse group that has remained recalcitrant to phylogenetic resolution. We combined 1007 exons obtained with Anchored Hybrid Enrichment with 1048 ultra-conserved elements (UCEs) for 433 taxa including all extant families, >95% of all subfamilies, and 356 genera chosen to represent the vast diversity of the superfamily. Going back and forth between the molecular results and our collective knowledge of morphology and biology, we detected bias in the analyses that was driven by the saturation of nucleotide data. Our final results are based on a concatenated analysis of the least saturated exons and UCE datasets (2054 loci, 284 106 sites). Our analyses support an expected sister relationship with Mymarommatoidea. Seven previously recognized families were not monophyletic, so support for a new classification is discussed. Natural history in some cases would appear to be more informative than morphology, as illustrated by the elucidation of a clade of plant gall associates and a clade of taxa with planidial first-instar larvae. The phylogeny suggests a transition from smaller soft-bodied wasps to larger and more heavily sclerotized wasps, with egg parasitism as potentially ancestral for the entire superfamily. Deep divergences in Chalcidoidea coincide with an increase in insect families in the fossil record, and an early shift to phytophagy corresponds with the beginning of the "Angiosperm Terrestrial Revolution". Our dating analyses suggest a middle Jurassic origin of 174 Ma (167.3-180.5 Ma) and a crown age of 162.2 Ma (153.9-169.8 Ma) for Chalcidoidea. During the Cretaceous, Chalcidoidea may have undergone a rapid radiation in southern Gondwana with subsequent dispersals to the Northern Hemisphere. This scenario is discussed with regard to knowledge about the host taxa of chalcid wasps, their fossil record and Earth's palaeogeographic history.

Sulforaphene suppresses oesophageal cancer growth through mitogen- and stress-activated kinase 2 in a PDX mouse model.

BACKGROUND: Although sulforaphene has potential anticancer effects, little is known about its effect on oesophageal squamous cell carcinoma (ESCC) invasiveness. METHODS: To investigate whether sulforaphene inhibits the growth of oesophageal cancer cells, MTT and anchorage-independent cell growth assays were performed. Global changes in the proteome and phosphoproteome of oesophageal cancer cells after sulforaphene treatment were analysed by mass spectrometry (MS), and the underlying molecular mechanism was further verified by in vivo and in vitro experiments. RESULTS: Sulforaphene treatment markedly affected proteins that regulate several cellular processes in oesophageal cancer cells, and mitogen- and stress-activated kinase 2 (MSK2) was the main genetic target of sulforaphene in reducing the growth of oesophageal cancer cells. Sulforaphene significantly suppressed ESCC cell proliferation in vitro and reduced the tumour size in an oesophageal patient-derived xenograft (PDX) SCID mouse model. Furthermore, the binding of sulforaphane to MSK2 in vitro was verified using a cellular thermal dhift assay, and the effect of MSK2 knockdown on the ESCC phenotype was observed using a shMSK2 model. CONCLUSION: The results showed that sulforaphene suppresses ESCC growth in both human oesophageal squamous cells and PDX mouse model by inhibiting MSK2 expression, implicating sulforaphene as a promising candidate for ESCC treatment.

Stress hyperglycemia is associated with disease severity in COVID-19.

INTRODUCTION: Coronavirus disease 2019 (COVID-19) is a global pandemic that has affected millions of people worldwide. In this paper, we analyse the relationship between stress hyperglycaemia and disease severity in patients with COVID-19. MATERIAL AND METHODS: A total of 252 patients with COVID-19 were included in this study. The patients were divided into the following groups: COVID-19 with stress hyperglycaemia (SHG), COVID-19 with diabetes (DM), and COVID-19 with normal blood glucose (NG). The stress hyperglycaemia rate (SHR) was calculated using the fasting blood glucose (FBG)/glycated haemoglobin (HbA1c) ratio. To further compare the disease characteristics of different SHRs, we divided the SHR into low SHR and high SHR according to the SHR median. Correlations between the severity of the disease and other factors were analysed after adjusting for sex and age. Multivariate analysis was performed using logistic regression to analyse the risk factors predicting the severity of COVID-19. RESULTS: Compared with the NG group, the SHG group had higher disease severity (p < 0.001); the SHG group had higher HbA1c, FBG, SHR, blood urea nitrogen (BUN), interleukin 6 (IL-6), and neutrophil levels, while lymphocyte, CD3+ T cell, CD8+ T cell, CD4+ T cell, CD16+CD56 cell, and CD19+ cell counts were lower (p < 0.05). Compared with the NG group, the DM group had higher HbA1c, blood glucose, BUN, lactate dehydrogenase (LDH), erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), and neutrophils, while CD8+ T cell counts were lower (p < 0.05). Compared with the DM group, the SHG group had higher SHR and lower HbA1c, CD3+ T cell, CD4+ T cell, CD16+CD56 cell, and T cell ratio levels (p < 0.05). Compared to the low SHR group, the high SHR group had patients with more severe COVID-19 (p = 0.004). Also, the high SHR grouphad higher age, HbA1c, FBG, asparate aminotransferaze (AST), BUN, LDH, uric acid (UA), CRP, IL-6, and procalcitonin (PCT), while lymphocyte, CD3+ T cell, CD4+ T cell, CD8+ T cell, and CD19+ cell counts were lower (p < 0.05).Binary logistic regression analysis showed that SHR, gender, and lymphocyte count wererisk factorsfor the severity of COVID-19. CONCLUSION: Stress hyperglycaemia, as indicated by a higher SHR, is independently associated with the severity of COVID-19.

Hypoxanthine phosphoribosyl transferase 1 metabolizes temozolomide to activate AMPK for driving chemoresistance of glioblastomas.

Temozolomide (TMZ) is a standard treatment for glioblastoma (GBM) patients. However, TMZ has moderate therapeutic effects due to chemoresistance of GBM cells through less clarified mechanisms. Here, we demonstrate that TMZ-derived 5-aminoimidazole-4-carboxamide (AICA) is converted to AICA ribosyl-5-phosphate (AICAR) in GBM cells. This conversion is catalyzed by hypoxanthine phosphoribosyl transferase 1 (HPRT1), which is highly expressed in human GBMs. As the bona fide activator of AMP-activated protein kinase (AMPK), TMZ-derived AICAR activates AMPK to phosphorylate threonine 52 (T52) of RRM1, the catalytic subunit of ribonucleotide reductase (RNR), leading to RNR activation and increased production of dNTPs to fuel the repairment of TMZ-induced-DNA damage. RRM1 T52A expression, genetic interruption of HPRT1-mediated AICAR production, or administration of 6-mercaptopurine (6-MP), a clinically approved inhibitor of HPRT1, blocks TMZ-induced AMPK activation and sensitizes brain tumor cells to TMZ treatment in mice. In addition, HPRT1 expression levels are positively correlated with poor prognosis in GBM patients who received TMZ treatment. These results uncover a critical bifunctional role of TMZ in GBM treatment that leads to chemoresistance. Our findings underscore the potential of combined administration of clinically available 6-MP to overcome TMZ chemoresistance and improve GBM treatment.

Noninvasive radiomics model reveals macrophage infiltration in glioma.

Preoperative MRI is an essential diagnostic and therapeutic reference for gliomas. This study aims to evaluate the prognostic aspect of a radiomics biomarker for glioma and further investigate its relationship with tumor microenvironment and macrophage infiltration. We covered preoperative MRI of 664 glioma patients from three independent datasets: Jiangsu Province Hospital (JSPH, n = 338), The Cancer Genome Atlas dataset (TCGA, n = 252), and Repository of Molecular Brain Neoplasia Data (REMBRANDT, n = 74). Incorporating a multistep post-processing workflow, 20 radiomics features (Rads) were selected and a radiomics survival biomarker (RadSurv) was developed, proving highly efficient in risk stratification of gliomas (cut-off = 1.06), as well as lower-grade gliomas (cut-off = 0.64) and glioblastomas (cut-off = 1.80) through three fixed cut-off values. Through immune infiltration analysis, we found a positive correlation between RadSurv and macrophage infiltration (RMΦ = 0.297, p < 0.001; RM2Φ = 0.241, p < 0.001), further confirmed by immunohistochemical-staining (glioblastomas, n = 32) and single-cell sequencing (multifocal glioblastomas, n = 2). In conclusion, RadSurv acts as a strong prognostic biomarker for gliomas, exhibiting a non-negligible positive correlation with macrophage infiltration, especially with M2 macrophage, which strongly suggests the promise of radiomics-based models as a preoperative alternative to conventional genomics for predicting tumor macrophage infiltration and provides clinical guidance for immunotherapy.

Deep-learning and conventional radiomics to predict IDH genotyping status based on magnetic resonance imaging data in adult diffuse glioma.

Objectives: In adult diffuse glioma, preoperative detection of isocitrate dehydrogenase (IDH) status helps clinicians develop surgical strategies and evaluate patient prognosis. Here, we aim to identify an optimal machine-learning model for prediction of IDH genotyping by combining deep-learning (DL) signatures and conventional radiomics (CR) features as model predictors. Methods: In this study, a total of 486 patients with adult diffuse gliomas were retrospectively collected from our medical center (n=268) and the public database (TCGA, n=218). All included patients were randomly divided into the training and validation sets by using nested 10-fold cross-validation. A total of 6,736 CR features were extracted from four MRI modalities in each patient, namely T1WI, T1CE, T2WI, and FLAIR. The LASSO algorithm was performed for CR feature selection. In each MRI modality, we applied a CNN+LSTM-based neural network to extract DL features and integrate these features into a DL signature after the fully connected layer with sigmoid activation. Eight classic machine-learning models were analyzed and compared in terms of their prediction performance and stability in IDH genotyping by combining the LASSO-selected CR features and integrated DL signatures as model predictors. In the validation sets, the prediction performance was evaluated by using accuracy and the area under the curve (AUC) of the receiver operating characteristics, while the model stability was analyzed by using the relative standard deviation of the AUC (RSDAUC). Subgroup analyses of DL signatures and CR features were also individually conducted to explore their independent prediction values. Results: Logistic regression (LR) achieved favorable prediction performance (AUC: 0.920 ± 0.043, accuracy: 0.843 ± 0.044), whereas support vector machine with the linear kernel (l-SVM) displayed low prediction performance (AUC: 0.812 ± 0.052, accuracy: 0.821 ± 0.050). With regard to stability, LR also showed high robustness against data perturbation (RSDAUC: 4.7%). Subgroup analyses showed that DL signatures outperformed CR features (DL, AUC: 0.915 ± 0.054, accuracy: 0.835 ± 0.061, RSDAUC: 5.9%; CR, AUC: 0.830 ± 0.066, accuracy: 0.771 ± 0.051, RSDAUC: 8.0%), while DL and DL+CR achieved similar prediction results. Conclusion: In IDH genotyping, LR is a promising machine-learning classification model. Compared with CR features, DL signatures exhibit markedly superior prediction values and discriminative capability.