Machine learning predictors of risk of death within 7 days in patients with non-traumatic subarachnoid hemorrhage in the intensive care unit: A multicenter retrospective study.

Non-traumatic subarachnoid hemorrhage (SAH) is a critical neurosurgical emergency with a high mortality rate, imposing a significant burden on both society and families. Accurate prediction of the risk of death within 7 days in SAH patients can provide valuable information for clinicians, enabling them to make better-informed medical decisions. In this study, we developed six machine learning models using the MIMIC III database and data collected at our institution. These models include Logistic Regression (LR), AdaBoosting (AB), Multilayer Perceptron (MLP), Bagging (BAG), Gradient Boosting Machines (GBM), and Extreme Gradient Boosting (XGB). The primary objective was to identify predictors of death within 7 days in SAH patients admitted to intensive care units. We employed univariate and multivariate logistic regression as well as Pearson correlation analysis to screen the clinical variables of the patients. The initially screened variables were then incorporated into the machine learning models, and the performance of these models was evaluated. Furthermore, we compared the performance differences among the six models and found that the MLP model exhibited the highest performance with an AUC of 0.913. In this study, we conducted risk factor analysis using Shapley values to identify the factors associated with death within 7 days in patients with SAH. The risk factors we identified include Gcsmotor, bicarbonate, wbc, spo2, heartrate, age, nely, glucose, aniongap, GCS, rbc, sysbp, sodium, and gcseys. To provide clinicians with a useful tool for assessing the risk of death within 7 days in SAH patients, we developed a web calculator based on the MLP machine learning model.

CSF Biopsy in Glioma: A Brief Review.

Glioma is the most common intracranial malignant tumor. Over the past several years, liquid biopsy in diagnosis and treatment of solid tumors have made many progressions, but there is still a gap from a large clinical application of liquid biopsy in glioma due to many limitations. However, in recent years, researchers have made many explorations into liquid biopsy in glioma. In the future, the liquid biopsy of glioma, especially cerebrospinal fluid, will have a broad prospect. In this review, we will discuss the current research progressions of CSF biopsy in glioma in recent years.

Machine learning-based prediction of cerebral hemorrhage in patients with hemodialysis: A multicenter, retrospective study.

Background: Intracerebral hemorrhage (ICH) is one of the most serious complications in patients with chronic kidney disease undergoing long-term hemodialysis. It has high mortality and disability rates and imposes a serious economic burden on the patient's family and society. An early prediction of ICH is essential for timely intervention and improving prognosis. This study aims to build an interpretable machine learning-based model to predict the risk of ICH in patients undergoing hemodialysis. Methods: The clinical data of 393 patients with end-stage kidney disease undergoing hemodialysis at three different centers between August 2014 and August 2022 were retrospectively analyzed. A total of 70% of the samples were randomly selected as the training set, and the remaining 30% were used as the validation set. Five machine learning (ML) algorithms, namely, support vector machine (SVM), extreme gradient boosting (XGB), complement Naïve Bayes (CNB), K-nearest neighbor (KNN), and logistic regression (LR), were used to develop a model to predict the risk of ICH in patients with uremia undergoing long-term hemodialysis. In addition, the area under the curve (AUC) values were evaluated to compare the performance of each algorithmic model. Global and individual interpretive analyses of the model were performed using importance ranking and Shapley additive explanations (SHAP) in the training set. Results: A total of 73 patients undergoing hemodialysis developed spontaneous ICH among the 393 patients included in the study. The AUC of SVM, CNB, KNN, LR, and XGB models in the validation dataset were 0.725 (95% CI: 0.610 ~ 0.841), 0.797 (95% CI: 0.690 ~ 0.905), 0.675 (95% CI: 0.560 ~ 0.789), 0.922 (95% CI: 0.862 ~ 0.981), and 0.979 (95% CI: 0.953 ~ 1.000), respectively. Therefore, the XGBoost model had the best performance among the five algorithms. SHAP analysis revealed that the levels of LDL, HDL, CRP, and HGB and pre-hemodialysis blood pressure were the most important factors. Conclusion: The XGB model developed in this study can efficiently predict the risk of a cerebral hemorrhage in patients with uremia undergoing long-term hemodialysis and can help clinicians to make more individualized and rational clinical decisions. ICH events in patients undergoing maintenance hemodialysis (MHD) are associated with serum LDL, HDL, CRP, HGB, and pre-hemodialysis SBP levels.

KIF4A drives gliomas growth by transcriptional repression of Rac1/Cdc42 to induce cytoskeletal remodeling in glioma cells.

Glioma is one of the most prevalent cancers diseases in the worldwide. Kinesin superfamily protein 4 (KIF4), a KIF member classified in Kinesin 4 has been indicated as a mediator acted in tumorigenesis of human cancer. However, the mechanism of KIF4A on glioma is yet to be investigated. This study aimed to explore the potential function and mechanism of KIF4A in gliomas. We analyzed the KIF4A expression and the prognosis in gliomas patients using The Cancer Genome Atlas (TCGA) databases. KIF4A level in normal human astrocyte cell (NHA) and glioma cell lines were examined by Western blot. We studied the function of KIF4A on proliferation, migration, invasion, cell cycle in glioma cell lines using a series of in vitro and in vivo experiments. Chromatin Immunoprecipitation (ChIP) analysis was applied to searching potential KIF4A related downstream in glioma. We identified the significant up-regulated expression of KIF4A both in glioma tissues and cell. Glioma patients with elevated KIF4A expression have shorter survival. Down-regulation of KIF4A exerted inhibitory effect on cell proliferation, invasion and migration. We crucially identified that KIF4A drives gliomas growth by transcriptional repression of Rac1/Cdc42 to induce cytoskeletal remodeling in glioma cells. Knockdown of KIF4A decreased RohA, Rac1, Cdc42, Pak1 and Pak2 expression level. Our study provided a prospect that KIF4A functions as an oncogene in glioma.

Vitamin C through upregulating SYNPO2 level suppresses the proliferation and migration of glioma cells.

PURPOSE: The aim was to demonstrate the progression of glioma influenced by Vitamin C (VC) and the potential molecular mechanism. METHODS: Proliferative and migratory rates of U87 and U251 cells induced with 0, 50 and 100 µM VC were examined by CCK-8 and Transwell assay, respectively. Clinical significance of SYNPO2 in glioma patients was analyzed. Relative level of SYNPO2 in VC-induced glioma cells was detected. By intervening SYNPO2, the involvement of SYNPO2 in the anti-cancer role of VC in inhibiting glioma cell phenotypes was finally confirmed. RESULTS: VC induction dose-dependently attenuated proliferative and migratory potentials of glioma cells. A low level of SYNPO2 indicated poor prognosis of glioma. Protein and mRNA levels of SYNPO2 were upregulated in glioma cells induced with VC. The inhibitory effects of VC on proliferative and migratory potentials of glioma cells were partially reversed by knockdown of SYNPO2. CONCLUSIONS: VC blocks glioma cells to proliferate and migrate by upregulating SYNPO2.

Rap2B promotes cell adhesion, proliferation, migration and invasion of human glioma.

PURPOSE: Rap2B, a member of the GTP-binding proteins, is generally up-regulated in numerous types of tumors. Nevertheless, the influence and regulatory mechanisms of Rap2B in gliomas are still not corroborated. Therefore, we analyzed the expression of Rap2B in glioma tissues and cells, and researched its significance in adhesion, proliferation, migration and invasion of the glioma cell line. METHODS: We analyzed the expression of Rap2B in different pathologic grades of glioma tissues by tissue microarray and immunohistochemistry. We assessed the expression of Rap2B in glioma tissue and non-tumor tissue by Western blot. And the expression of Rap2b protein in glioma cells and normal human astrocytes (NHA) was detected by Western blot. In addition, we disclosed the effect of Rap2B knockdown on cell adhesion, proliferation, migration and invasion by using cell attachment assay, CCK-8 assay, cell migration assay and Wound Healing assay, cell invasion assay, respectively. Western blot was used to detect the changes of expression level of NF-kB, MMP-2 and MMP-9 protein when downregulated the expression of Rap2B. RESULTS: The tissue microarray immunohistochemical results of glioma showed that the expression of Rap2B had no significant correlations between Rap2B expression and the clinicopathologic variables, including patient age (P = 0.352), gender (P = 0.858), WHO Grade (P = 0.693) and histology type (P = 0.877). Western blot analysis showed that the glioma tissue had a dramatically increase of Rap2B expression compared with the non-tumor tissues (P < 0.01). And the expression of Rap2B was markedly up-regulated in all 5 glioma cell lines compared with that in normal human astrocytes (NHA) (P < 0.01). We found that the ability of adhesion, proliferation, migration and invasion of glioma cells were significantly decreased after downregulated Rap2B expression compared with the control group (P < 0.05). In addition, Western blot results showed that the expression levels of NF-kB, MMP-2 and MMP-9 in the interference group were significantly lower than those in the negative control group (P < 0.05). CONCLUSIONS: Rap2B expression is up-regulated in glioma tissues and glioma cell lines. Knockdown of Rap2B inhibits glioma cells' adhesion and proliferation in vitro. Knockdown of Rap2B inhibits glioma cells' migration in vitro. Knockdown of Rap2B inhibits glioma cells' invasion and MMPs activity through NF-kB pathway.

Relationship between expression of XRCC1 and tumor proliferation, migration, invasion, and angiogenesis in glioma.

Recently, XRCC1 polymorphisms were reported to be associated with glioma in Chinese population. However, only a few studies reported on the XRCC1 expression, and cancer progression. In this study, we investigated whether XRCC1 plays a role in glioma pathogenesis. Using the tissue microarray technology, we found that XRCC1 expression is significantly decreased in glioma compared with tumor adjacent normal brain tissue (P < 0.01, χ2 test) and reduced XRCC1 staining was associated with WHO stages (P < 0.05, χ2 test). The mRNA and protein levels of XRCC1 were significantly downregulated in human primary glioma tissues (P < 0.001, χ2 test). We also found that XRCC1 was significantly decreased in glioma cell lines compared to normal human astrocytes (P < 0.01, χ2 test). Overexpression of XRCC1 dramatically reduced the proliferation and caused cessation of cell cycle. The reduced cell proliferation is due to G1 phase arrest as cyclin D1 is diminished whereas p16 is upregulated. We further demonstrated that XRCC1 overexpression suppressed the glioma cell migration and invasion abilities by targeting MMP-2. In addition, we also found that overexpression of XRCC1 sharply inhibited angiogenesis, which correlated with down-regulation of VEGF. The data indicate that XRCC1 may be a tumor suppressor involved in the progression of glioma.

CTHRC1 mediates multiple pathways regulating cell invasion, migration and adhesion in glioma.

Recently, collagen triple helix repeat containing-1 (CTHRC1) has been reported to be increased in several types of human solid cancers and to be associated with tumor invasion and metastasis. However, the expression and function of CTHRC1 in glioma have not yet been reported. In the present study, we investigated whether CTHRC1 plays a role in glioma pathogenesis. Using the tissue microarray technology, we found that CTHRC1 expression is significantly increased in glioma compared with tumor adjacent normal brain tissue (P<0.01, χ2 test) and increased CTHRC1 staining was associated with WHO stages (P<0.05, χ2 test). The mRNA and protein levels of CTHRC1 were significantly upregulated in human primary glioma tissues (P<0.001, χ2 test). We also found that CTHRC1 was significantly increased in glioma cell lines compared to normal human astrocytes (P<0.01, χ2 test). Furthermore, Knockdown of CTHRC1 suppressed glioma cell invasion and inhibited enzyme activity of MMP-2. Moreover, our data showed that knockdown of CTHRC1 inhibited glioma cell migration and adhesion capacity when compared with the control cells, and CTHRC1-siRNA reduced the levels of phosphorylated Src and FAK protein expression. Taken together, this study suggests that CTHRC1 plays a role in glioma development and progression by regulating invasion, migration and adhesion capabilities of cancer cells.

Overexpression of CAP1 and its significance in tumor cell proliferation, migration and invasion in glioma.

Adenylate cyclase-associated protein 1 (CAP1), a protein related to the regulation of actin filaments and the Ras/cAMP pathway, is associated with tumor progression. Nevertheless, the expression level and effects of CAP1 in regards to glioma have not been reported. In the present study, we examined the expression of CAP1 in glioma and tumor adjacent normal brain tissues by tissue microarray and immunohistochemistry. Our results showed that CAP1 was overexpressed in glioma tissues in comparison with that noted in the tumor adjacent normal brain tissues and increased staining of CAP1 was found to be correlated with WHO stage. In addition, we discovered that knockdown of CAP1 by specific RNA interference markedly inhibited cell growth and caused downregulation of the proliferation markers, PCNA and cyclin A. We further demonstrated that knockdown of CAP1 inhibited cell metastatic abilities by downregulating N-cadherin and vimentin and upregulating E-cadherin. These findings revealed that CAP1 expression is markedly increased in human glioma and that downregulation of CAP1 in tumors may serve as a treatment for glioma patients.

Cullin1 regulates proliferation, migration and invasion of glioma cells.

This study was designed to explore the role of Cullin1 (Cul1) in the pathogenesis of human glioma and to investigate the role of Cul1 in the growth, migration and invasion of glioma cells. Expression of Cul1 in 191 glioma tissues, 8 normal brain tissues and 8 tumor adjacent normal brain tissues was analyzed by tissue microarray and immunohistochemistry. Cul1 expression in human glioblastoma cells was knocked down by specific siRNA to study the effect of down-regulation of Cul1 on proliferation, invasion and migration of glioma cells. Our results showed that Cul1 expression increased significantly in tissues from the benign tumor and malignant tumor in comparison with those from the tumor-adjacent normal brain (P<0.05 for both). We did not find any correlation between Cul1 expression and clinicopathological parameters. In addition, we found that knockdown of Cul1 by RNA interference markedly inhibited cell proliferation and caused cessation of cell cycle. This reduced cell proliferation was due to G1 phase arrest as cyclinA, cyclinD1 and cyclinE were diminished, whereas p21 and p27 were up-regulated. We further demonstrated that silencing of Cul1 in glioma cells inhibited the cell migration and invasion abilities, and down-regulation of MMP-2 and MMP-9 expression greatly contributed to the reduced cell invasion and migration abilities. Our data indicated that Cul1 expression significantly increased in human glioma, and it may be involved in proliferation, migration and invasion of glioma cells.

BRMS1 suppresses glioma progression by regulating invasion, migration and adhesion of glioma cells.

Breast cancer metastasis suppressor 1 (BRMS1) is a metastasis suppressor gene in several solid tumors. However, the expression and function of BRMS1 in glioma have not been reported. In this study, we investigated whether BRMS1 play a role in glioma pathogenesis. Using the tissue microarray technology, we found that BRMS1 expression is significantly decreased in glioma compared with tumor adjacent normal brain tissue (P<0.01, χ(2) test) and reduced BRMS1 staining is associated with WHO stages (P<0.05, χ(2) test). We also found that BRMS1 was significantly downregulated in glioma cell lines compared to normal human astrocytes (P<0.01, χ(2) test). Furthermore, we demonstrated that BRMS1 overexpression inhibited glioma cell invasion by suppressing uPA, NF-κB, MMP-2 expression and MMP-2 enzyme activity. Moreover, our data showed that overexpression of BRMS1 inhibited glioma cell migration and adhesion capacity compared with the control group through the Src-FAK pathway. Taken together, this study suggested that BRMS1 has a role in glioma development and progression by regulating invasion, migration and adhesion activities of cancer cells.

MiR-29c inhibits glioma cell proliferation, migration, invasion and angiogenesis.

Previous studies reported that miR-29c is significantly downregulated in several tumors. However, little is known about the effect and molecular mechanisms of action of miR-29c in human glioma. Using quantitative RT-PCR, we demonstrated that miR-29c was significantly downregulated in glioma cell lines and human primary glioma tissues, compared to normal human astrocytes and matched non-tumor associated tissues (P < 0.05, χ(2) test). Overexpression of miR-29c dramatically reduced the proliferation and caused cessation of cell cycle. The reduced cell proliferation is due to G1 phase arrest as cyclin D1 and cyclin E are diminished whereas p27 and p21 are upregulated. We further demonstrated that miR-29c overexpression suppressed the glioma cell migration and invasion abilities by targeting MMP-2. In addition, we also found that overexpression of miR-29c sharply inhibited angiogenesis, which correlated with down-regulation of VEGF. The data indicate that miR-29c may be a tumor suppressor involved in the progression of glioma.

[Application of diffusion tensor imaging in preoperation and postoperation patients of glioma with 3.0 Tesla MRI].

OBJECTIVE: To investigate the clinical usefulness of diffusion tensor imaging (DTI) in demonstrating between gliomas and surrounding fibers. METHODS: 24 patients of glioma (WHO grade: grade I - II 16 cases; grade III - IV 8 cases) were examined using DTI and conventional contrast-enhanced MRI of 3.0 T MRI scanner (GE company, America). After the initial data acquisition introduced into workstation, image analysis was performed with the use of functool software. Mean diffusivity (MD) and fractional anisotropy (FA) values were measured in regions of solid tumor, surrounding edema and normal white matter of the high grade cerebral gliomas. Differences in these values among the tissues were assessed on the high grade cerebral gliomas. Anatomic relationship between intracranial tumors and surrounding fibers was analysed on fractional anisotropic (FA) map, color-coded directional map, three-dimensional white matter tractography. All patients' symptoms were evaluated preoperative and postoperative respectively. RESULTS: The DTI patterns altered by the tumor were categorized as follows: displacement, infiltration and destruction. The tractography showed that the main influence in 16 cases of grade I - II glioma on adjacent white matter tracts was displacement, but infiltration and destruction were also revealed. The patients have definite improvement in symptoms. The destruction and infiltration of fiber tracts could be seen in all edema regions around grade III - IV gliomas in 8 cases. The patients have not definite improvement in symptoms. Apparently significant differences of MD were found in solid tumor, surrounding edema, compared with normal white matter regions (P < 0.05). But there was no significant difference among solid tumor and surrounding edema region (P > 0.05). There were significant differences of FA between solid tumor, surrounding edema and normal white matter region (P < 0.05). CONCLUSIONS: The DTI offered the optimal visualization of white matter tracts. DTI plays an important role in demonstrating relationship between gliomas and neighboring fibers. MD and FA values could be used to distinguish normal white matter from solid tumor and surrounding edema region of high grade glioma. The application of DTI in preoperation plays an guidance role in making microsurgery plans and the evaluation of brain functional recovery in postoperation. DTI should be of great value in the microsurgical planning as well as estimation and reduction of potential postoperative neurological deficits for the cerebral gliomas resection.