Improving the Accuracy and Efficiency of Abnormal Cervical Squamous Cell Detection With Cytologist-in-the-Loop Artificial Intelligence.

Population-based cervical cytology screening techniques are demanding and laborious and have relatively poor diagnostic accuracy. In this study, we present a cytologist-in-the-loop artificial intelligence (CITL-AI) system to improve the accuracy and efficiency of abnormal cervical squamous cell detection in cervical cancer screening. The artificial intelligence (AI) system was developed using 8000 digitalized whole slide images, including 5713 negative and 2287 positive cases. External validation was performed using an independent, multicenter, real-world data set of 3514 women, who were screened for cervical cancer between 2021 and 2022. Each slide was assessed using the AI system, which generated risk scores. These scores were then used to optimize the triaging of true negative cases. The remaining slides were interpreted by cytologists who had varying degrees of experience and were categorized as either junior or senior specialists. Stand-alone AI had a sensitivity of 89.4% and a specificity of 66.4%. These data points were used to establish the lowest AI-based risk score (ie, 0.35) to optimize the triage configuration. A total of 1319 slides were triaged without missing any abnormal squamous cases. This also reduced the cytology workload by 37.5%. Reader analysis found CITL-AI had superior sensitivity and specificity compared with junior cytologists (81.6% vs 53.1% and 78.9% vs 66.2%, respectively; both with P < .001). For senior cytologists, CITL-AI specificity increased slightly from 89.9% to 91.5% (P = .029); however, sensitivity did not significantly increase (P = .450). Therefore, CITL-AI could reduce cytologists' workload by more than one-third while simultaneously improving diagnostic accuracy, especially compared with less experienced cytologists. This approach could improve the accuracy and efficiency of abnormal cervical squamous cell detection in cervical cancer screening programs worldwide.

ARL4C stabilized by AKT/mTOR pathway promotes the invasion of PTEN-deficient primary human glioblastoma.

Phosphatase and tensin homolog deleted on chromosome 10 (PTEN) deficiency in primary human glioblastoma (GBM) is associated with increased invasiveness and poor prognosis with unknown mechanisms. Therefore, how loss of PTEN promotes GBM progression remains to be elucidated. Herein, we identified that ADP-ribosylation factor like-4C (ARL4C) was highly expressed in PTEN-deficient human GBM cells and tissues. Mechanistically, loss of PTEN stabilized ARL4C protein due to AKT/mTOR pathway-mediated inhibition of ARL4C ubiquitination. Functionally, ARL4C enhanced the progression of GBM cells in vitro and in vivo. Moreover, microarray profiling and GST pull-down assay identified that ARL4C accelerated tumor progression via RAC1-mediated filopodium formation. Importantly, targeting PTEN potently inhibited GBM tumor progression in vitro and in vivo, whereas overexpression of ARL4C reversed the tumor progression impaired by PTEN overexpression. Clinically, analyses with patients' specimens validated a negative correlation between PTEN and ARL4C expression. Elevated ARL4C expression but PTEN deficiency in tumor was associated with poorer disease-free survival and overall survival of GBM patients. Taken together, ARL4C is critical for PTEN-deficient GBM progression and acts as a novel prognostic biomarker and a potential therapeutic candidate. Copyright © 2018 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.

Identification of a Gene-Related Risk Signature in Melanoma Patients Using Bioinformatic Profiling.

INTRODUCTION: Gene signature has been used to predict prognosis in melanoma patients. Meanwhile, the efficacy of immunotherapy was correlated with particular genes expression or mutation. In this study, we systematically explored the gene expression pattern in the melanoma-immune microenvironment and its relationship with prognosis. METHODS: A cohort of 122 melanoma cases with whole-genome microarray expression data were enrolled from the Gene Expression Omnibus (GEO) database. The findings were validated using The Cancer Genome Atlas (TCGA) database. A principal component analysis (PCA), gene set enrichment analysis (GSEA), and gene oncology (GO) analysis were performed to explore the bioinformatic implications. RESULTS: Different gene expression patterns were identified according to the clinical stage. All eligible gene sets were analyzed, and the 8 genes (GPR87, KIT, SH3GL3, PVRL1, ATP1B1, CDAN1, FAU, and TNFSF14) with the greatest prognostic impact on melanoma. A gene-related risk signature was developed to distinguish patients with a high or low risk of an unfavorable outcome, and this signature was validated using the TCGA database. Furthermore, the prognostic significance of the signature between the classified subgroups was verified as an independent prognostic predictor of melanoma. Additionally, the low-risk melanoma patients presented an enhanced immune phenotype compared to that of the high-risk gene signature patients. CONCLUSIONS: The gene pattern differences in melanoma were profiled, and a gene signature that could independently predict melanoma patients with a high risk of poor survival was established, highlighting the relationship between prognosis and the local immune response.

Autophagy-induced KDR/VEGFR-2 activation promotes the formation of vasculogenic mimicry by glioma stem cells.

Antiangiogenesis with bevacizumab, an antibody against vascular endothelial growth factor (VEGF), has been used for devascularization to limit the growth of malignant glioma. However, the benefits are transient due to elusive mechanisms underlying resistance to the antiangiogenic therapy. Glioma stem cells (GSCs) are capable of forming vasculogenic mimicry (VM), an alternative microvascular circulation independent of VEGF-driven angiogenesis. Herein, we report that the formation of VM was promoted by bevacizumab-induced macroautophagy/autophagy in GSCs, which was associated with tumor resistance to antiangiogenic therapy. We established a 3-dimensional collagen scaffold to examine the formation of VM and autophagy by GSCs, and found that rapamycin increased the number of VM and enhanced KDR/VEGFR-2 phosphorylation. Treatment with chloroquine, or knockdown of the autophagy gene ATG5, inhibited the formation of VM and KDR phosphorylation in GSCs. Notably, neutralization of GSCs-produced VEGF with bevacizumab failed to recapitulate the effect of chloroquine treatment and ATG5 knockdown, suggesting that autophagy-promoted formation of VM was independent of tumor cell-derived VEGF. ROS was elevated when autophagy was induced in GSCs and activated KDR phosphorylation through the phosphoinositide 3-kinase (PI3K)-AKT pathway. A ROS inhibitor, N-acetylcysteine, abolished KDR phosphorylation and the formation of VM by GSCs. By examination of the specimens from 95 patients with glioblastoma, we found that ATG5 and p-KDR expression was strongly associated with the density of VM in tumors and poor clinical outcome. Our results thus demonstrate a crucial role of autophagy in the formation of VM by GSCs, which may serve as a therapeutic target in drug-resistant glioma.

[Expression and Clinical Pathological Significance of EBER, PTEN and VEGF in Angioimmunoblastic T-Cell Lymphoma].

OBJECTIVE: To investigate the expression and clinical pathological significance of EB virus (Epstein-Barr virus, EBV), PTEN and VEGF in angioimmunoblastic T -cell lymphoma (AITL). METHODS: The EBV -encoded small RNA (EBER) expression in 21 cases of AITL was detected by in situ hybridization. The expressions of PTEN and VEGF were detected in 21 cases of AITL and 20 cases of lymph node reactive hyperplasia by immunohistochemical EnVision two-steps method. The expression and clinicopathological significance of EBV, PTEN and VEGF in AITL were analyzed. RESULTS: The positive expression rate of EBER in 21 cases of AITL was 61.9%; the expressions of PTEN and VEGF in AITL and lymph node reactive hyperplasia were significantly different (P<0.05). The expressions of EBER and PTEN negatively correlated (P<0.05). The EBER positive expression rates of male patients in AITL group and the progressed group was 80% and 78.6% respectively, which were significantly higher than that in female patients and patients in non- advanced group (P<0.05); the PTEN expression rates in the AITL group accompanying B symptoms and progressed group were 31.3% and 21.4%, respectively, which were significantly lower than those in patients without B symptoms and non-progressed group (P<0.05). Survival analysis showed that the PTEN expression negatively correlated with the overall survival rate of patients (P<0.05). CONCLUSION: EBV infection and low expression of PTEN may indicate the deterioration of angioimmunoblastic T-cell lymphoma. Whether the EBV involved in the ocurring of T-cell angioimmunoblastic lymphoma by down-regulating PTEN expression is unclear, further research is needed.