Stacking Ensemble Learning-Based [18F]FDG PET Radiomics for Outcome Prediction in Diffuse Large B-Cell Lymphoma.

This study aimed to develop an analytic approach based on [18F]FDG PET radiomics using stacking ensemble learning to improve the outcome prediction in diffuse large B-cell lymphoma (DLBCL). Methods: In total, 240 DLBCL patients from 2 medical centers were divided into the training set (n = 141), internal testing set (n = 61), and external testing set (n = 38). Radiomics features were extracted from pretreatment [18F]FDG PET scans at the patient level using 4 semiautomatic segmentation methods (SUV threshold of 2.5, SUV threshold of 4.0 [SUV4.0], 41% of SUVmax, and SUV threshold of mean liver uptake [PERCIST]). All extracted features were harmonized with the ComBat method. The intraclass correlation coefficient was used to evaluate the reliability of radiomics features extracted by different segmentation methods. Features from the most reliable segmentation method were selected by Pearson correlation coefficient analysis and the LASSO (least absolute shrinkage and selection operator) algorithm. A stacking ensemble learning approach was applied to build radiomics-only and combined clinical-radiomics models for prediction of 2-y progression-free survival and overall survival based on 4 machine learning classifiers (support vector machine, random forests, gradient boosting decision tree, and adaptive boosting). Confusion matrix, receiver-operating-characteristic curve analysis, and survival analysis were used to evaluate the model performance. Results: Among 4 semiautomatic segmentation methods, SUV4.0 segmentation yielded the highest interobserver reliability, with 830 (66.7%) selected radiomics features. The combined model constructed by the stacking method achieved the best discrimination performance. For progression-free survival prediction in the external testing set, the areas under the receiver-operating-characteristic curve and accuracy of the stacking-based combined model were 0.771 and 0.789, respectively. For overall survival prediction, the stacking-based combined model achieved an area under the curve of 0.725 and an accuracy of 0.763 in the external testing set. The combined model also demonstrated a more distinct risk stratification than the International Prognostic Index in all sets (log-rank test, all P < 0.05). Conclusion: The combined model that incorporates [18F]FDG PET radiomics and clinical characteristics based on stacking ensemble learning could enable improved risk stratification in DLBCL.

Therapeutic potential of tumor treating fields for malignant brain tumors.

BACKGROUND: Malignant brain tumors are among the most threatening diseases of the central nervous system, and despite increasingly updated treatments, the prognosis has not been improved. Tumor treating fields (TTFields) are an emerging approach in cancer treatment using intermediate-frequency and low-intensity electric field and can lead to the development of novel therapeutic options. RECENT FINDINGS: A series of biological processes induced by TTFields to exert anti-cancer effects have been identified. Recent studies have shown that TTFields can alter the bioelectrical state of macromolecules and organelles involved in cancer biology. Massive alterations in cancer cell proteomics and transcriptomics caused by TTFields were related to cell biological processes as well as multiple organelle structures and activities. This review addresses the mechanisms of TTFields and recent advances in the application of TTFields therapy in malignant brain tumors, especially in glioblastoma (GBM). CONCLUSIONS: As a novel therapeutic strategy, TTFields have shown promising results in many clinical trials, especially in GBM, and continue to evolve. A growing number of patients with malignant brain tumors are being enrolled in ongoing clinical studies demonstrating that TTFields-based combination therapies can improve treatment outcomes.

Optical molecular imaging and theranostics in neurological diseases based on aggregation-induced emission luminogens.

Optical molecular imaging and image-guided theranostics benefit from special and specific imaging agents, for which aggregation-induced emission luminogens (AIEgens) have been regarded as good candidates in many biomedical applications. They display a large Stokes shift, high quantum yield, good biocompatibility, and resistance to photobleaching. Neurological diseases are becoming a substantial burden on individuals and society that affect over 50 million people worldwide. It is urgently needed to explore in more detail the brain structure and function, learn more about pathological processes of neurological diseases, and develop more efficient approaches for theranostics. Many AIEgens have been successfully designed, synthesized, and further applied for molecular imaging and image-guided theranostics in neurological diseases such as cerebrovascular disease, neurodegenerative disease, and brain tumor, which help us understand more about the pathophysiological state of brain through noninvasive optical imaging approaches. Herein, we focus on representative AIEgens investigated on brain vasculature imaging and theranostics in neurological diseases including cerebrovascular disease, neurodegenerative disease, and brain tumor. Considering different imaging modalities and various therapeutic functions, AIEgens have great potential to broaden neurological research and meet urgent needs in clinical practice. It will be inspiring to develop more practical and versatile AIEgens as molecular imaging agents for preclinical and clinical use on neurological diseases.

Fine particulate matter (PM2.5) promotes IgE-mediated mast cell activation through ROS/Gadd45b/JNK axis.

BACKGROUND: Mast cells play an important role in allergic responses and persistently exposure to environmental fine particulate matter (PM2.5) exacerbates allergic diseases,but the details remained elucidative. OBJECTIVES: To investigate the effect of PM2.5 on IgE-mediated mast cell responses through an IgE-mediated mouse model and mast cell activation. METHODS: The ß-hexosaminidase release and a BALB/c model of passive cutaneous anaphylaxis (PCA) was used to test IgE-mediated mast cells activation in vitro and in vivo. RNA-Seq technique was conducted to study the gene expression profile. Reactive oxygen species (ROS) production was measured by flow-cytometry. RT-PCR,WB and ELISA were performed to examine targeting molecules expression. RESULTS: PM2.5 facilitated IgE-mediated degranulation and increased cytokines expression in mast cells. Meanwhile, the Evan's blue extravasation as well as serum cytokines in mice was increased after treatment with PM2.5. Furthermore, PM2.5 treatment dramatically increased the expression of Gadd45b which is an oxidative stress molecule that directly activates down-stream pathway, such as MEKK4/JNK. PM2.5 treatment activated MEKK4, JNK1/2 but not ERK1/2 and p38. Meanwhile, Knockdown of Gadd45b significantly attenuated PM2.5-mediated JNK1/2 activation and expression of cytokines. In addition, a JNK1/2-specific inhibitor SP600125 blocked IgE-mediated mast cell activation and cytokine release in PCA model mice. Moreover, PM2.5 treatment increased the ROS level and ROS inhibitor dramatically blocked the PM2.5-induced ROS production and reversed the PM2.5-mediated gene expression in the mitochondrial respiratory chain. CONCLUSIONS: PM2.5 regulates ROS production through Gadd45b/MEKK4/JNK pathway, facilitating IgE-mediated mast cell activation.

Cloning of the XPD gene and its function in malignant melanoma cells.

The xeroderma pigmentosum group D (XPD) gene is a member of the transcription factor IIH complex and serves an important role in gene repair. Previous studies have suggested that genetic variants of the XPD gene may be associated with an increased risk of cutaneous melanoma. However, the exact mechanism remains unclear. In the present study, the XPD gene was cloned, and its localization and function in malignant melanoma cells were investigated. The human full length XPD gene was cloned via reverse transcription-PCR using the total RNA extracted from human cervical squamous cell carcinoma epithelial HeLa cells. Subsequently, the gene was inserted into a plasmid fused to green fluorescent protein (GFP; pEGFP-N1/XPD), and pEGFP-N1/XPD and pcDNA3.1(+)/XPD were transfected into human malignant melanoma A375 cells using Lipofectamine® 2000. The expression levels of XPD were detected by western blotting. The Golgi marker GM130 and the endoplasmic reticulum membrane protein marker KDEL were used for immunofluorescence staining, and the subcellular localization of XPD was observed under a fluorescence microscope. Cell proliferation was measured using an MTT assay. The recombinant pEGFP-N1/XPD plasmid expressing the human wild-type XPD gene was successfully constructed by restriction enzyme digestion and assessed by gene sequencing. XPD was localized in the endoplasmic reticulum of malignant melanoma A375 cells, as confirmed by immunofluorescence staining. Furthermore, MTT assays indicated that XPD inhibited the proliferation of malignant melanoma A375 cells. The present study provides a basis for further investigation of the biological effects and functions of XPD in malignant melanoma cells.

Benserazide is a novel inhibitor targeting PKM2 for melanoma treatment.

The M2 splice isoform of pyruvate kinase (PKM2) is a key enzyme for generating pyruvate and ATP in the glycolytic pathway, whereas the role of PKM2 in tumorigenesis remains a subject of debate. In our study, we found PKM2 is highly expressed in melanoma patients and the malignance is positively correlated with high PKM2 activity and glycolytic capability in melanoma cells. Suppression of PKM2 expression by knocking down markedly attenuated malignant phenotype both in vitro and in vivo, and restoration of PKM2 expression in PKM2 depleted cells could rescue melanoma cells proliferation, invasion and metastasis. With the data indicating PKM2 as a potential therapeutic target, we performed screening for PKM2 inhibitors and identified benserazide (Ben), a drug currently in clinical use. We demonstrated that Ben directly binds to and blocks PKM2 enzyme activity, leading to inhibition of aerobic glycolysis concurrent up-regulation of OXPHOS. Of note, despite PKM2 is very similar to PKM1, Ben does not affect PKM1 enzyme activity. We showed that Ben significantly inhibits cell proliferation, colony formation, invasion and migration in vitro and in vivo. The specificity of Ben was demonstrated by the findings that, suppression of PKM2 expression diminishes the efficacy of Ben in inhibition of melanoma cell growth; ectopic PKM2 expression in normal cells sensitizes cells to Ben treatment. Interestingly, PKM2 activity and aerobic glycolysis are upregulated in BRAFi-resistant melanoma cells. As a result, BRAFi-resistant cells exhibit heightened sensitivity to suppression of PKM2 expression or treatment with Ben both in vitro and in vivo.

Integration of microRNAome, proteomics and metabolomics to analyze arsenic-induced malignant cell transformation.

Long-term exposure to arsenic has been linked to tumorigenesis in different organs and tissues, such as skin; however, the detailed mechanism remains unclear. In this present study, we integrated "omics" including microRNAome, proteomics and metabolomics to investigate the potential molecular mechanisms. Compared with non-malignant human keratinocytes (HaCaT), twenty-six miRNAs were significantly altered in arsenic-induced transformed cells. Among these miRNAs, the differential expression of six miRNAs was confirmed using Q-RT-PCR, representing potential oxidative stress genes. Two-dimensional gel electrophoresis (2D-PAGE) and mass spectrometry (MS) were performed to identify the differential expression of proteins in arsenic-induced transformed cells, and twelve proteins were significantly changed. Several proteins were associated with oxidative stress and carcinogenesis including heat shock protein beta-1 (HSPB1), peroxiredoxin-2 (PRDX2). Using ultra-performance liquid chromatography and Q-TOF mass spectrometry (UPLC/Q-TOF MS), 68 metabolites including glutathione, fumaric acid, citric acid, phenylalanine, and tyrosine, related to redox metabolism, glutathione metabolism, citrate cycle, met cycle, phenylalanine and tyrosine metabolism were identified and quantified. Taken together, these results indicated that arsenic-induced transformed cells exhibit alterations in miRNA, protein and metabolite profiles providing novel insights into arsenic-induced cell malignant transformation and identifying early potential biomarkers for cutaneous squamous cell carcinoma induced by arsenic.

Repressing CD147 is a novel therapeutic strategy for malignant melanoma.

CD147/basigin, a transmembrane protein, is a member of the immunoglobulin super family. Accumulating evidence has revealed the role of CD147 in the development and progression of various cancers, including malignant melanoma (MM). MM is a malignancy of pigment-producing cells that causes the greatest number of skin cancer-related deaths worldwide. CD147 is overexpressed in MM and plays an important role in cell viability, apoptosis, proliferation, invasion, and metastasis, probably by mediating vascular endothelial growth factor (VEGF) production, glycolysis, and multi-drug resistance (MDR). As a matrix metalloproteinase (MMP) inducer, CD147 could also promote surrounding fibroblasts to secrete abundant MMPs to further stimulate tumor cell invasion. Targeting CD147 has been shown to suppress MM in vitro and in vivo, highlighting the therapeutic potential of CD147 silencing in MM treatment. In this review article, we discuss CD147 and its biological roles, regulatory mechanisms, and potential application as a molecular target for MM.

Epigallocatechin-3-gallate(EGCG) suppresses melanoma cell growth and metastasis by targeting TRAF6 activity.

TRAF6 (TNF Receptor-Associated Factor 6) is an E3 ubiquitin ligase that contains a Ring domain, induces K63-linked polyubiquitination, and plays a critical role in signaling transduction. Our previous results demonstrated that TRAF6 is overexpressed in melanoma and that TRAF6 knockdown dramatically attenuates tumor cell growth and metastasis. In this study, we found that EGCG can directly bind to TRAF6, and a computational model of the interaction between EGCG and TRAF6 revealed that EGCG probably interacts with TRAF6 at the residues of Gln54, Gly55, Asp57 ILe72, Cys73 and Lys96. Among these amino acids, mutation of Gln54, Asp57, ILe72 in TRAF6 could destroy EGCG bound to TRAF6, furthermore, our results demonstrated that EGCG significantly attenuates interaction between TRAF6 and UBC13(E2) and suppresses TRAF6 E3 ubiquitin ligase activity in vivo and in vitro. Additionally, the phosphorylation of IκBα, p-TAK1 expression are decreased and the nuclear translocation of p65 and p50 is blocked by treatment with EGCG, leading to inactivation of the NF-κB pathway. Moreover, EGCG significantly inhibits cell growth as well as the migration and invasion of melanoma cells. Taken together, these findings show that EGCG is a novel E3 ubiquitin ligase inhibitor that could be used to target TRAF6 for chemotherapy or the prevention of melanoma.

Novel CFTR gene mutation in a patient with CBAVD.

We report a novel mutation detected in a 33 year old Chinese man with congenital bilateral absence of the vas deferens (CBAVD), a past history of pulmonary meliodosis infection and a past history of bronchiolitis obliterans organising pneumonia. A novel splice site mutation in intron 6b (1001+5 G-->A) in the homozygous state was identified, and was predicted to lead to inefficient splicing. He was also homozygous at all intragenic and flanking polymorphic markers. Quantitative realtime PCR analysis showed that there were 2 copies of the CFTR gene present, ruling out the possibility of a deletion, and strongly suggesting the possibility of uniparental isodisomy involving at least a part of chromosome 7.