A Hybrid Deep Learning Model for Predicting Molecular Subtypes of Human Breast Cancer Using Multimodal Data

BackgroundThe prediction of breast cancer subtypes plays a key role in the diagnosis and prognosis of breast cancer. In recent years, deep learning (DL) has shown good performance in the intelligent prediction of breast cancer subtypes. However, most of the traditional DL models use single modality data, which can just extract a few features, so it cannot establish a stable relationship between patient characteristics and breast cancer subtypes.DatasetWe used the TCGA-BRCA dataset as a sample set for molecular subtype prediction of breast cancer. It is a public dataset that can be obtained through the following link: https://portal.gdc.cancer.gov/projects/TCGA-BRCAMethodsIn this paper, a Hybrid DL model based on the multimodal data is proposed. We combine the patient's gene modality data with image modality data to construct a multimodal fusion framework. According to the different forms and states, we set up feature extraction networks respectively, and then we fuse the output of the two feature networks based on the idea of weighted linear aggregation. Finally, the fused features are used to predict breast cancer subtypes. In particular, we use the principal component analysis to reduce the dimensionality of high-dimensional data of gene modality and filter the data of image modality. Besides, we also improve the traditional feature extraction network to make it show better performance.ResultsThe results show that compared with the traditional DL model, the Hybrid DL model proposed in this paper is more accurate and efficient in predicting breast cancer subtypes. Our model achieved a prediction accuracy of 88.07% in 10 times of 10-fold cross-validation. We did a separate AUC test for each subtype, and the average AUC value obtained was 0.9427. In terms of subtype prediction accuracy, our model is about 7.45% higher than the previous average.     

The role of microbicidal lipids in host defense against pathogens and their potential as therapeutic agents

Lipids such as fatty alcohols, free fatty acids and monoglycerides of fatty acids are known to be potent antimicrobial/microbicidal agents in vitro and to kill enveloped viruses, Gram-positive and Gram-negative bacteria and fungi on contact. For over half a century several studies have tried to answer the question of whether or not lipids play a role in the natural host defense against pathogens. A comprehensive review is given of these studies, particularly concerning infections in skin and in mucosal membranes of the respiratory tract, and of the role of lipids in the antimicrobial activity of breast milk. Based on studies of the microbicidal activities of lipids, both in vitro and in vivo, the possibility of using such lipids as active ingredients in prophylactic and therapeutic dosage forms is considered and examples are given of studies of such pharmaceutical dosage forms in experimental animal models and in clinical trials.     

CpG/CpNpG motifs in the coding region are preferred sites for mutagenesis in the breast cancer susceptibility genes

The range of BRCA1/BRCA2 gene mutations is diverse and the mechanism accounting for this heterogeneity is obscure. To gain insight into the endogenous mutational mechanisms involved, we evaluated the association of specific sequences (i.e. CpG/CpNpG motifs, homonucleotides, short repeats) and mutations within the genes. We classified 1337 published mutations in BRCA1 (1765 BRCA2 mutations) for each specific sequence, and employed computer simulation combined with mathematical calculations to estimate the true underlying tendency of mutation occurrence. Interestingly, we found no mutational bias to homonucleotides and repeats in deletions/insertions and substitutions but striking bias to CpG/CpNpG in substitutions in both genes. This suggests that methylation-dependent DNA alterations would be a major mechanism for mutagenesis.     

希罗达对4T1乳腺癌小鼠肿瘤血管生成的影响

目的:探讨不同剂量希罗达对4T1乳腺癌小鼠肿瘤血管生成的影响,并观察其抑瘤效果和毒副作用。方法:建立荷4T1乳腺癌小鼠模型,分别给予持续低剂量希罗达、最大耐受剂量(maximumtolerateddose,MTD)希罗达、低剂量希罗达治疗,观察肿瘤体积、小鼠体重和外周血白细胞计数及小鼠生存期。实验终末时行免疫组化染色,测定肿瘤微血管密度(MVD)和血管内皮生长因子(VEGF)表达情况。结果:与对照组和MTD希罗达治疗组比较,持续低剂量希罗达治疗组小鼠肿瘤MVD值和VEGF表达均显著降低(P<0.05)。与MTD希罗达治疗组比较,持续低剂量希罗达治疗组肿瘤生长比较缓慢,并且没有明显的体重减轻或白细胞数下降等毒性迹象,小鼠生存期无显著延长(P>0.05)。结论:持续低剂量希罗达给药方式靶向于乳腺癌血管生成,不易产生耐药,增加了抑瘤效果,毒副作用小,动物生存质量明显提高。     

C—erbB2与乳腺癌三苯氧胺耐药细胞生长关系的研究

目的：探讨获得性三苯氧胺（TAM）抵抗乳腺癌细胞的生长调节途径及三苯氧胺（TAM）获得性抵抗的发生机制。方法：TAM诱导野生型人乳腺癌细胞系MCF-7/WT构建TAM抵抗的细胞系MCF-7／TAMR,RT—PCR、Westem blot及免疫细胞化学方法比较MCF-7／TAMR与MCF-7/WT细胞系中C—erbB2mRNA、蛋白表达及其活化状态的不同,用C—erbB2单克隆抗体herceptin对两种细胞系进行干预。观察细胞生长变化。结果：与MCF-7/WT细胞相比,MCF-7／TAMR细胞中cerbB-2的mRNA增加3倍（P〈0．05）,蛋白增加1．5倍（P〈0．05）。Herceptin处理MCF-7／TAMR细胞,明显抑制了细胞的生长。结论：表皮生长因子受体特异性配体的自分泌释放作用可能通过CerbB-2／MAPK途径引起MCF-7／TAMR细胞的生长。     

Role of the Wnt and GTPase pathways in breast cancer tumorigenesis and treatment

Breast cancer (BC) is the most frequently diagnosed cancer among women in all the populations of the world. Although the BC mortality rate has declined, resistance to treatment is still a significant challenge for patient survival. Various cellular signaling pathways, such as Wnt and Rho/GTPase have been linked to the development, migration, and metastasis of BC, and also in treatment resistance mechanisms. Some studies have shown an association between two important cellular pathways, Wnt and Rho/GTPase, in cytoskeleton activation and cancer invasion. However, their involvement in BC has received little attention. This review summarizes the Wnt and Rho/GTPases signaling pathway functions, and also the crosstalk between these pathways in the progression, metastasis, and drug resistance mechanisms in BC. Considering the signaling pathways involved in BC tumorigenesis, future studies will need to investigate possible molecular interventions and new opportunities for the development of personalized strategies for BC treatment in order to improve overall outcomes.     

钼靶和超声检查在乳腺癌临床诊断的准确性的比较分析

摘要 目的：比较与分析钼靶和超声检查在乳腺癌临床诊断的准确性。方法：2018年8月到2021年1月选择在本院进行诊治的乳腺肿瘤患者110例作为研究对象,所有患者都给予钼靶和超声检查,记录影像学特征并判断诊断价值。结果：在110例患者中,病理诊断为乳腺良性肿瘤76例、乳腺癌34例。恶性组钼靶的分叶征、钙化、大角征、毛刺征等比例高于良性组,病灶大小也高于良性组(P<0.05)。恶性组超声的形态不规则、边缘不光整、高回声晕、回声衰减、微钙化等比例高于良性组(P<0.05)。钼靶乳腺影像报告及数据系统(Breast imaging report and data system,BI-RADS)判断为乳腺良性肿瘤72例,乳腺癌38例;超声BI-RADS判断为乳腺良性肿瘤75例,乳腺癌35例,钼靶鉴别诊断乳腺癌的敏感性为93.4%,特异性为97.1%,准确性为94.5%;超声鉴别诊断乳腺癌的敏感性为98.7%,特异性为100.0%,准确性为99.1%。多因素logistic回归分析显示病灶大小、分叶征、回声衰减、毛刺征为导致误诊的重要因素(P<0.05)。结论：乳腺癌在钼靶和超声检查中都有明显的征象特征,超声诊断的准确性更高,病灶大小、分叶征、回声衰减、毛刺征为影响诊断效果的很重要因素。     

乳腺癌术后患者创伤后成长与心理弹性和生命质量的相关性研究

摘要 目的：探讨乳腺癌术后患者创伤后成长（PTG）的影响因素,并分析其与心理弹性和生命质量的关系。方法：选择2020年1月~2021年8月在山东省立第三医院接受手术治疗的乳腺癌患者80例作为研究对象。采用创伤后成长量表（PTGI）评价乳腺癌术后患者PTG情况。采用乳腺癌患者生命质量测定量表（FACT-B）评价生命质量。采用心理弹性量表（CD-RISC）评价心理弹性。采用单因素及多因素Logistic回归分析乳腺癌术后患者PTG的影响因素,采用Pearson相关性分析PTGI评分与CD-RISC评分和FACT-B评分的关系。结果：乳腺癌术后患者PTGI评分为（50.38±13.39）分,根据PTGI评分将其分为PTG高分组（n=32）和PTG低分组（n=48）。乳腺癌术后患者PTG水平与文化程度、家庭月收入、疾病发现方式、工作状况、手术方式、临床分期、术后并发症、基础疾病有关（P<0.05）。多因素Logistic回归分析结果显示：家庭月收入<3000元、临床分期为II期、文化程度为小学及其以下、手术方式为乳房切除、有术后并发症、有基础疾病均是乳腺癌术后患者PTG的危险因素（P<0.05）。PTG高分组患者的CD-RISC、FACT-B评分高于PTG低分组,组间对比有统计学差异（P<0.05）。Pearson相关性分析结果显示,PTGI评分与CD-RISC评分、FACT-B评分呈正相关（P<0.05）。结论：乳腺癌术后患者PTG水平一般,受到临床分期、文化程度、手术方式等多种因素的影响,临床工作中应对相关因素予以重视并及时干预。同时PTG会影响患者的心理弹性和生命质量,PTG水平越高,心理弹性越好,生命质量亦越高。     

An aptasensor for rapid and sensitive detection of estrogen receptor alpha in human breast cancer

The analysis of estrogen receptor (ER) expression in breast carcinomas plays a crucial role in determining the endocrine responsiveness of tumors for systemic adjuvant therapy. Conventionally, the ER levels in breast carcinomas had been detected using the dextran-coated charcoal assay and radioimmunoassay, which are now substituted with safer and economic antibody-based assays such as immunohistochemistry (IHC) and enzyme-linked immunosorbent assay (ELISA). Despite a gold (Au) standard method, the IHC has been criticized for factors such as tissue fixation, antibody selection, and threshold staining for result interpretation that could falsify test accuracy and reproducibility. The quest for alternative methods of ER quantification in tissue samples paved the way for aptamer-based diagnostics. Previously, we have isolated a DNA aptamer against human ER alpha (ERα) using an in vitro evolution system. In this study, we developed an electrochemical sensor using the 76-nucleotide DNA ERα- aptamer for rapid, precise, and cost-effective detection of ERα expression in human breast cancer patients. The aptasensor was constructed by covalently immobilizing the thiolated ERα- aptamer onto a screen-printed Au electrode. Construction of aptasensors was confirmed through atomic force microscopy and differential pulse voltammetry measurements. A detection limit of 0.001 ng/ml was calculated for full-length ERα (66.2 kDa) in a detection time of 10 min. Analysis of the cancerous breast tissue samples using the ELISA and aptasensor methods enabled distinctive classification of samples into the categories of ER −ve, weak ER +ve, and strong ER +ve samples. The current change of this aptasensor lies within 5% after a storage of 60 days at 4°C. Further studies on a reasonably large sample size are required to realize the clinical potential of the sensor.