AGR3 in breast cancer: prognostic impact and suitable serum-based biomarker for early cancer detection

High resolution mapping of the tumor microenvironment using integrated single-cell, spatial and in situ analysis

Single-cell and spatial technologies that profile gene expression across a whole tissue are revolutionizing the resolution of molecular states in clinical samples. Current commercially available technologies provide whole transcriptome single-cell, whole transcriptome spatial, or targeted in situ gene expression analysis. Here, we combine these technologies to explore tissue heterogeneity in large, FFPE human breast cancer sections. This integrative approach allowed us to explore molecular differences that exist between distinct tumor regions and to identify biomarkers involved in the progression towards invasive carcinoma. Further, we study cell neighborhoods and identify rare boundary cells that sit at the critical myoepithelial border confining the spread of malignant cells. Here, we demonstrate that each technology alone provides information about molecular signatures relevant to understanding cancer heterogeneity; however, it is the integration of these technologies that leads to deeper insights, ushering in discoveries that will progress oncology research and the development of diagnostics and therapeutics.

Breast cancer in young Indian women: factors, challenges in screening, and upcoming diagnostics

Breast cancer management for young Indian women are full of challenges. The National Cancer Registry Programme (NCRP) has predicted that nearly 2,30,000 cases of breast cancer will be reported annually by 2025; with a steady increase in cases of young women (< 45 years of age) with breast cancer. In this review, the available literature is evaluated to understand the various risk factors contributing to the rise in cases of breast cancer in young women in India. Further, the challenges that are faced by the technicians in early diagnosis (e.g., physiology of young breasts, limited trained professionals, and awareness among patients, and cost of the treatment) of breast cancer. This review also focuses on the upcoming diagnostics like serum biomarkers and nanosensors for the early identification of the disease. For better prognosis and to reduce the chances of disease reoccurrence and metastasis, it is important that the disease has to be identified at an early stage.

Prognostic Impact of AGR3 Protein Expression in Breast Cancer: A Systematic Review and Meta-analysis

OBJECTIVE

 To investigate the clinicopathological significance and prognosis of the expression of the anterior gradient 3 (AGR3) protein in women with breast cancer.

DATA SOURCES

 The PubMed, CINAHL, EMBASE, Scopus, and Web of Science databases were searched for studies published in English and without restrictions regarding the year of publication. The search terms were: breast cancer AND anterior gradient 3 OR AGR3 expression.

STUDY SELECTION

 We included observational or interventional studies, studies on AGR3 protein expression by immunohistochemistry, and studies on invasive breast cancer. Case reports, studies with animals, and reviews were excluded. In total, 4 studies were included, containing 713 cases of breast cancer.

DATA COLLECTION

 Data were extracted on clinicopathological characteristics and survival. A meta-analysis of the prevalence of AGR3 expression was performed according to the clinicopathological characteristics, hazard ratios (HRs), and overall survival and disease-free survival.

DATA SYNTHESIS

 The expression of AGR3 was found in 62% of the cases, and it was associated with histological grade II, positivity of estrogen and progesterone receptors, low expression of ki67, recurrence or distant metastasis, and lumen subtypes. In patients with low and intermediate histological grades, AGR3 expression was associated with worse overall survival (HR: 2.39; 95% confidence interval [95%CI]: 0.628-4.159; p = 0.008) and worse disease-free survival (HR: 3.856; 95%CI: 1.026-6.686; p = 0.008).

CONCLUSION

 The AGR3 protein may be a biomarker for the early detection of breast cancer and predict prognosis in luminal subtypes. In addition, in patients with low and intermediate histological grades, AGR3 protein expression may indicate an unfavorable prognosis in relation to survival.

Breast cancer cell secretome analysis to decipher miRNA regulating the tumor microenvironment and discover potential biomarkers

MicroRNA (miRNA/miR) 526 b- and miR655-overexpressed tumor cell-free secretions regulate the breast cancer tumor microenvironment (TME) by promoting tumor-associated angiogenesis, oxidative stress, and hypoxic responses. Additionally, premature miRNA (pri-miR526b and pri-miR655) are established breast cancer blood biomarkers. However, the mechanisms of how these miRNAs regulate the TME has yet to be investigated. Mass spectrometry analysis of miRNA-overexpressed cell lines MCF7-miR526b, MCF7-miR655, and miRNA-low MCF7-Mock cell-free secretomes identified 34 differentially expressed proteins coded by eight genes. In both miRNA-high cell secretomes, four markers are upregulated: YWHAB, SFN, TXNDC12, and MYL6B, and four are downregulated: PEA15, PRDX4, PSMB6, and FN1. All upregulated marker transcripts are significantly high in both total cellular RNA pool and cell-free secretions of miRNA-high cell lines, validated with quantitative RT-PCR. Bioinformatics tools were used to investigate these markers' roles in breast cancer. These markers' top gene ontology functions are related to apoptosis, oxidative stress, membrane transport, and motility supporting oncogenic miR526b- and miR655-induced functions. Gene transcription factor analysis tools were used to show how these miRNAs regulate the expression of each secretory marker. Data extracted from the Human Protein Atlas showed that YWHAB, SFN, and TXNDC12 expression could distinguish early and late-stage breast cancer in various breast cancer subtypes and are associated with poor patient survival. Additionally, immunohistochemistry analysis showed the expression of each marker in breast tumors. A stronger correlation between miRNA clusters and upregulated secretory markers gene expression was found in the luminal A tumor subtype. YWHAB, SFN, and MYL6B are upregulated in breast cancer patient's blood, showing biomarker potential. Of these identified novel miRNA secretory markers, SFN and YWHAB successfully passed all validations and are the best candidates to further investigate their roles in miRNA associated TME regulation. Also, these markers show the potential to serve as blood-based breast cancer biomarkers, especially for luminal-A subtypes.

Identification of a gene expression signature associated with breast cancer survival and risk that improves clinical genomic platforms

Motivation

Modern genomic technologies allow us to perform genome-wide analysis to find gene markers associated with the risk and survival in cancer patients. Accurate risk prediction and patient stratification based on robust gene signatures is a key path forward in personalized treatment and precision medicine. Several authors have proposed the identification of gene signatures to assign risk in patients with breast cancer (BRCA), and some of these signatures have been implemented within commercial platforms in the clinic, such as Oncotype and Prosigna. However, these platforms are black boxes in which the influence of selected genes as survival markers is unclear and where the risk scores provided cannot be clearly related to the standard clinicopathological tumor markers obtained by immunohistochemistry (IHC), which guide clinical and therapeutic decisions in breast cancer.

Results

Here, we present a framework to discover a robust list of gene expression markers associated with survival that can be biologically interpreted in terms of the three main biomolecular factors (IHC clinical markers: ER, PR and HER2) that define clinical outcome in BRCA. To test and ensure the reproducibility of the results, we compiled and analyzed two independent datasets with a large number of tumor samples (1024 and 879) that include full genome-wide expression profiles and survival data. Using these two cohorts, we obtained a robust subset of gene survival markers that correlate well with the major IHC clinical markers used in breast cancer. The geneset of survival markers that we identify (which includes 34 genes) significantly improves the risk prediction provided by the genesets included in the commercial platforms: Oncotype (16 genes) and Prosigna (50 genes, i.e. PAM50). Furthermore, some of the genes identified have recently been proposed in the literature as new prognostic markers and may deserve more attention in current clinical trials to improve breast cancer risk prediction.

Availability and implementation

All data integrated and analyzed in this research will be available on GitHub (https://github.com/jdelasrivas-lab/breastcancersurvsign), including the R scripts and protocols used for the analyses.

Supplementary information

Supplementary data are available at Bioinformatics Advances online.

Identifying gene expression-based biomarkers in online learning environments

Motivation

Gene expression-based classifiers are often developed using historical data by training a model on a small set of patients and a large set of features. Models trained in such a way can be afterwards applied for predicting the output for new unseen patient data. However, very often the accuracy of these models starts to decrease as soon as new data is fed into the trained model. This problem, known as concept drift, complicates the task of learning efficient biomarkers from data and requires special approaches, different from commonly used data mining techniques.

Results

Here, we propose an online ensemble learning method to continually validate and adjust gene expression-based biomarker panels over increasing volume of data. We also propose a computational solution to the problem of feature drift where gene expression signatures used to train the classifier become less relevant over time. A benchmark study was conducted to classify the breast tumors into known subtypes by using a large-scale transcriptomic dataset (∼3500 patients), which was obtained by combining two datasets: SCAN-B and TCGA-BRCA. Remarkably, the proposed strategy improves the classification performances of gold-standard biomarker panels (e.g. PAM50, OncotypeDX and Endopredict) by adding features that are clinically relevant. Moreover, test results show that newly discovered biomarker models can retain a high classification accuracy rate when changing the source generating the gene expression profiles.

Availability and implementation

github.com/UEFBiomedicalInformaticsLab/OnlineLearningBD.

Supplementary information

Supplementary data are available at Bioinformatics Advances online.

Functions and mechanisms of protein disulfide isomerase family in cancer emergence

The endoplasmic reticulum (ER) is a multi-layered organelle that is essential for the synthesis, folding, and structural maturation of almost one-third of the cellular proteome. It houses several resident proteins for these functions including the 21 members of the protein disulfide isomerase (PDI) family. The signature of proteins belonging to this family is the presence of the thioredoxin domain which mediates the formation, and rearrangement of disulfide bonds of substrate proteins in the ER. This process is crucial not only for the proper folding of ER substrates but also for maintaining a balanced ER proteostasis. The inclusion of new PDI members with a wide variety of structural determinants, size and enzymatic activity has brought additional epitomes of how PDI functions. Notably, some of them do not carry the thioredoxin domain and others have roles outside the ER. This also reflects that PDIs may have specialized functions and their functions are not limited within the ER. Large-scale expression datasets of human clinical samples have identified that the expression of PDI members is elevated in pathophysiological states like cancer. Subsequent functional interrogations using structural, molecular, cellular, and animal models suggest that some PDI members support the survival, progression, and metastasis of several cancer types. Herein, we review recent research advances on PDIs, vis-à-vis their expression, functions, and molecular mechanisms in supporting cancer growth with special emphasis on the anterior gradient (AGR) subfamily. Last, we posit the relevance and therapeutic strategies in targeting the PDIs in cancer.

Molecular Subtyping of Cancer Based on Robust Graph Neural Network and Multi-Omics Data Integration

Accurate molecular subtypes prediction of cancer patients is significant for personalized cancer diagnosis and treatments. Large amount of multi-omics data and the advancement of data-driven methods are expected to facilitate molecular subtyping of cancer. Most existing machine learning–based methods usually classify samples according to single omics data, fail to integrate multi-omics data to learn comprehensive representations of the samples, and ignore that information transfer and aggregation among samples can better represent them and ultimately help in classification. We propose a novel framework named multi-omics graph convolutional network (M-GCN) for molecular subtyping based on robust graph convolutional networks integrating multi-omics data. We first apply the Hilbert–Schmidt independence criterion least absolute shrinkage and selection operator (HSIC Lasso) to select the molecular subtype-related transcriptomic features and then construct a sample–sample similarity graph with low noise by using these features. Next, we take the selected gene expression, single nucleotide variants (SNV), and copy number variation (CNV) data as input and learn the multi-view representations of samples. On this basis, a robust variant of graph convolutional network (GCN) model is finally developed to obtain samples’ new representations by aggregating their subgraphs. Experimental results of breast and stomach cancer demonstrate that the classification performance of M-GCN is superior to other existing methods. Moreover, the identified subtype-specific biomarkers are highly consistent with current clinical understanding and promising to assist accurate diagnosis and targeted drug development.

Breast Cancer; Discovery of Novel Diagnostic Biomarkers, Drug Resistance, and Therapeutic Implications

Breast cancer is the second most reported cancer in women with high mortality causing millions of cancer-related deaths annually. Early detection of breast cancer intensifies the struggle towards discovering, developing, and optimizing diagnostic biomarkers that can improve its prognosis and therapeutic outcomes. Breast cancer-associated biomarkers comprise macromolecules, such as nucleic acid (DNA/RNA), proteins, and intact cells. Advancements in molecular technologies have identified all types of biomarkers that are exclusively studied for diagnostic, prognostic, drug resistance, and therapeutic implications. Identifying biomarkers may solve the problem of drug resistance which is a challenging obstacle in breast cancer treatment. Dysregulation of non-coding RNAs including circular RNAs (circRNAs) and microRNAs (miRNAs) initiates and progresses breast cancer. The circulating multiple miRNA profiles promise better diagnostic and prognostic performance and sensitivity than individual miRNAs. The high stability and existence of circRNAs in body fluids make them a promising new diagnostic biomarker. Many therapeutic-based novels targeting agents have been identified, including ESR1 mutation (DNA mutations), Oligonucleotide analogs and antagonists (miRNA), poly (ADP-ribose) polymerase (PARP) in BRCA mutations, CDK4/6 (cell cycle regulating factor initiates tumor progression), Androgen receptor (a steroid hormone receptor), that have entered clinical validation procedure. In this review, we summarize the role of novel breast cancer diagnostic biomarkers, drug resistance, and therapeutic implications for breast cancer.

Transcript-targeted analysis reveals isoform alterations and double-hop fusions in breast cancer

Although transcriptome alteration is an essential driver of carcinogenesis, the effects of chromosomal structural alterations on the cancer transcriptome are not yet fully understood. Short-read transcript sequencing has prevented researchers from directly exploring full-length transcripts, forcing them to focus on individual splice sites. Here, we develop a pipeline for Multi-Sample long-read Transcriptome Assembly (MuSTA), which enables construction of a transcriptome from long-read sequence data. Using the constructed transcriptome as a reference, we analyze RNA extracted from 22 clinical breast cancer specimens. We identify a comprehensive set of subtype-specific and differentially used isoforms, which extended our knowledge of isoform regulation to unannotated isoforms including a short form TNS3. We also find that the exon-intron structure of fusion transcripts depends on their genomic context, and we identify double-hop fusion transcripts that are transcribed from complex structural rearrangements. For example, a double-hop fusion results in aberrant expression of an endogenous retroviral gene, ERVFRD-1, which is normally expressed exclusively in placenta and is thought to protect fetus from maternal rejection; expression is elevated in several TCGA samples with ERVFRD-1 fusions. Our analyses provide direct evidence that full-length transcript sequencing of clinical samples can add to our understanding of cancer biology and genomics in general.

The Utility of SOX2 and AGR2 Biomarkers as Early Predictors of Tamoxifen Resistance in ER-Positive Breast Cancer Patients

BACKGROUND

Despite the undeniable benefit of tamoxifen therapy for ER-positive breast cancer patients, approximately one-third of those patients either do not respond to tamoxifen or develop resistance. Thus, it is a crucial step to identify novel, reliable, and easily detectable biomarkers indicating resistance to this drug.

OBJECTIVE

The aim of this work is to explore SOX2 and AGR2 biomarker expression in the tumor tissue of ER-positive breast cancer patients in combination with the evaluation of serum AGR2 level of these patients in order to validate these biomarkers as early predictors of tamoxifen resistance.

METHODS

This study was conducted on 224 ER-positive breast cancer patients. All patients were primarily subjected to serum AGR2 levelling by ELISA and their breast cancer tissue immunostained for SOX2 and AGR2. After 5 years of follow-up, the patients were divided into 3 groups: group 1 was tamoxifen sensitive and groups 2 and 3 were tamoxifen resistant. Time to failure of tamoxifen treatment was considered the time from the beginning of tamoxifen therapy to the time of discovery of breast cancer recurrence or metastases (in months).

RESULTS

SOX2 and AGR2 biomarkers expression and serum AGR2 level were significantly higher in groups 2 and 3 in comparison to group 1, while the relationship between Her2 neu expression and Ki67 index in the 3 different groups was statistically nonsignificant. Lower SOX2 and AGR2 expression and low AGR2 serum levels in the studied patients of groups 2 and 3 were significantly associated with longer time-to-failure of tamoxifen treatment. According to the ROC curve, the combined use of studied markers validity was with a sensitivity of 100%, specificity of 96%, PPV 96%, and NPV 100% (p < 0.001; AUC: 0.984).

CONCLUSIONS

Integrated use of SOX2 and AGR2 biomarkers with serum AGR2 assay holds a promising hope for their future use as predictive markers for early detection of tamoxifen resistance in ER-positive breast cancer patients.

Estrogen-regulated AGR3 activates the estrogen receptor signaling pathway to promote tamoxifen resistance in breast cancer

PURPOSE

Anterior gradient 3 (AGR3) is associated with breast cancer progression, but its relationship with estrogen and tamoxifen resistance in breast cancer is still unclear. This study was designed to investigate the correlation of ARG3 and estrogen as well as the roles of ARG3 in tamoxifen resistance in breast cancer.

METHODS

Online database including GEPIA, UALCAN, and TCGA and rVista predictive tool were applied to analyze the expression patterns of AGR3 and its relationship with estrogen receptor 1. AGR3 knockdown and overexpression cell models were constructed. Luciferase reporter assay and ChIP were performed to investigate intermolecular interactions. Western blotting and qPCR were applied to assess targets at mRNA and protein levels, respectively. Cell counting and MTT assay were applied to determine the cell proliferation.

RESULTS

An elevation of AGR3 was observed in patients with breast cancer, especially in the patients with estrogen receptor (ER)-positive breast cancer. The TCGA dataset and in vitro data supported that AGR3 was positively correlated to ER. Further results demonstrated that ER protein bound to AGR3 promoter sites. AGR3 expression exhibited a positive correlation to cell viability. Besides, AGR3 promoted tamoxifen resistance in breast cancer.

CONCLUSION

AGR3 is associated with estrogen and promotes tamoxifen resistance in breast cancer.

The Anterior GRadient (AGR) family proteins in epithelial ovarian cancer

Epithelial ovarian cancer (EOC) is the most common gynecologic disorder. Even with the recent progresses made towards the use of new therapeutics, it still represents the most lethal gynecologic malignancy in women from developed countries.

The discovery of the anterior gradient proteins AGR2 and AGR3, which are highly related members belonging to the protein disulfide isomerase (PDI) family, attracted researchers’ attention due to their putative involvement in adenocarcinoma development. This review compiles the current knowledge on the role of the AGR family and the expression of its members in EOC and discusses the potential clinical relevance of AGR2 and AGR3 for EOC diagnosis, prognosis, and therapeutics.

A better understanding of the role of the AGR family may thus provide new handling avenues for EOC patients.

AGR2 and AGR3 play an important role in the clinical characterization and prognosis of basal like breast cancer

INTRODUCTION

Basal-like Breast Cancer (BLBC) represents an important molecular subtype of breast cancer characterized by an aggressive behavior, molecular pathology poorly understood and a limited treatment.

OBJECTIVE

We aim to search for molecular differences between non-BLBC and BLBC tumors in order to propose possible diagnostic and prognostic biomarkers using databases. Metodology: Microarray processed data were downloaded from GEO database considering non-BLBC and BLBC. Enrichment analysis was evaluated using GO consortium and Ingenuity, protein-protein interaction, gene Ontology and co-expression analysis using STRING. Gene expression data was extracted using TCGA, METABRIC and Breast Cancer Gene-Expression Miner v4.2 databases. The Survival was evaluated using The Kaplan-Meier plotter.

RESULTS

Were identified 58 upregulated and 58 downregulated genes enriched in signaling pathways like PDGF, Angiogenesis, Integrin and WNT. AGR2 and AGR3 expression were reduced in BLBC in relation to non-BLBC tumors, patients aged ≤51 years, and with negativity of ER, PR and HER-2 and nodal status. Low expression of AGR2 and AGR3 were associated with worse OS and RFS for all breast cancer cases. But according to the molecular stratification, low AGR2 conferred worst OS in luminal A, worst RFS in BLBC and good OS and RFS in luminal B. High AGR3 conferred worse OS and RFS in BLBC, but low AGR3 attributed worse OS in luminal A.

CONCLUSION

AGR2 and AGR3 expression were able to differentiate non-BLBC from BLBC. Downregulation of AGR2 and AGR3 was associated with BLBC clinical phenotype. Furthermore, both genes behave different when considering prognosis and molecular stratification.

Profile of anterior gradient 3 (AGR3) mRNA expression and serum levels in benign and malignant breast tumors

BACKGROUND

Benign and malignant breast tumors are the most commonly diagnosed tumor in females. Early and accurate diagnosis of malignancy is essential for effective breast cancer treatment. Human anterior gradient 3 (AGR3) has been suggested as a potential biomarker for the early detection and prognostic determination of breast cancer.

OBJECTIVE

This study profiles AGR3 mRNA expression and serum protein levels in patients with benign and malignant breast tumors.

METHODS

A case-control study was conducted on 40 benign and 40 malignant breast tumor patients in Makassar, Indonesia. AGR3 mRNA and protein were detected using qRT-PCR and ELISA, respectively.

RESULTS

This study found significantly higher AGR3 mRNA expression in benign than malignant breast tumors using qRT-PCR (p < 0.001). In contrast, ELISA revealed no significant difference between AGR3 serum protein levels in benign and malignant breast tumors (p = 0.507).

CONCLUSIONS

AGR3 is associated with non-aggressive tumors and could be used as a marker for less aggressive breast tumors.

Anterior Gradient Protein 3 and S100 Calcium-Binding Protein P Levels in Different Endometrial Epithelial Compartments May Play an Important Role in Recurrent Pregnancy Failure

Recurrent implantation failure (RIF) and recurrent pregnancy loss (RPL) are distressing conditions without effective treatments. The luminal epithelium (LE) is integral in determining receptivity of the endometrium, whereas functionalis glands and stroma aid in nurturing early embryo development. Calcium signalling pathways are known to be of vital importance to embryo implantation and pregnancy establishment, and anterior gradient protein 3 (AGR3) and S100 calcium-binding protein P (S100P) are involved with these pathways. We initially examined 20 full-thickness endometrial biopsies from premenopausal women across the menstrual cycle to characterize levels of AGR3 protein in each endometrial sub-region at the cellular level. A further 53 endometrial pipelle biopsies collected in the window of implantation were subsequently assessed to determine differential endometrial AGR3 and S100P levels relevant to RIF (n = 13) and RPL (n = 10) in comparison with parous women (n = 30) using immunohistochemistry. Significantly higher AGR3 and S100P immunostaining was observed in ciliated cells of the LE of women with recurrent reproductive failure compared with parous women, suggesting aberrant subcellular location-associated pathophysiology for these conditions. The nuclear localisation of S100P may allow transcriptional regulatory function, which is necessary for implantation of a viable pregnancy. Further work is thus warranted to assess their utility as diagnostic/therapeutic targets.

Recent Discoveries of Macromolecule- and Cell-Based Biomarkers and Therapeutic Implications in Breast Cancer

Breast cancer is the most commonly diagnosed cancer type and the leading cause of cancer-related mortality in women worldwide. Breast cancer is fairly heterogeneous and reveals six molecular subtypes: luminal A, luminal B, HER2+, basal-like subtype (ER−, PR−, and HER2−), normal breast-like, and claudin-low. Breast cancer screening and early diagnosis play critical roles in improving therapeutic outcomes and prognosis. Mammography is currently the main commercially available detection method for breast cancer; however, it has numerous limitations. Therefore, reliable noninvasive diagnostic and prognostic biomarkers are required. Biomarkers used in cancer range from macromolecules, such as DNA, RNA, and proteins, to whole cells. Biomarkers for cancer risk, diagnosis, proliferation, metastasis, drug resistance, and prognosis have been identified in breast cancer. In addition, there is currently a greater demand for personalized or precise treatments; moreover, the identification of novel biomarkers to further the development of new drugs is urgently needed. In this review, we summarize and focus on the recent discoveries of promising macromolecules and cell-based biomarkers for the diagnosis and prognosis of breast cancer and provide implications for therapeutic strategies.

High-throughput proteomics of breast cancer interstitial fluid: identification of tumor subtype-specific serologically relevant biomarkers

Despite significant advancements in breast cancer (BC) research, clinicians lack robust serological protein markers for accurate diagnostics and tumor stratification. Tumor interstitial fluid (TIF) accumulates aberrantly externalized proteins within the local tumor space, which can potentially gain access to the circulatory system. As such, TIF may represent a valuable starting point for identifying relevant tumor-specific serological biomarkers. The aim of the study was to perform comprehensive proteomic profiling of TIF to identify proteins associated with BC tumor status and subtype. A liquid chromatography tandem mass spectrometry (LC-MS/MS) analysis of 35 TIFs of three main subtypes: luminal (19), Her2 (4), and triple-negative (TNBC) (12) resulted in the identification of > 8800 proteins. Unsupervised hierarchical clustering segregated the TIF proteome into two major clusters, luminal and TNBC/Her2 subgroups. High-grade tumors enriched with tumor infiltrating lymphocytes (TILs) were also stratified from low-grade tumors. A consensus analysis approach, including differential abundance analysis, selection operator regression, and random forest returned a minimal set of 24 proteins associated with BC subtypes, receptor status, and TIL scoring. Among them, a panel of 10 proteins, AGR3, BCAM, CELSR1, MIEN1, NAT1, PIP4K2B, SEC23B, THTPA, TMEM51, and ULBP2, was found to stratify the tumor subtype-specific TIFs. In particular, upregulation of BCAM and CELSR1 differentiates luminal subtypes, while upregulation of MIEN1 differentiates Her2 subtypes. Immunohistochemistry analysis showed a direct correlation between protein abundance in TIFs and intratumor expression levels for all 10 proteins. Sensitivity and specificity were estimated for this protein panel by using an independent, comprehensive breast tumor proteome dataset. The results of this analysis strongly support our data, with eight of the proteins potentially representing biomarkers for stratification of BC subtypes. Five of the most representative proteomics databases currently available were also used to estimate the potential for these selected proteins to serve as putative serological markers.

Integrative Generalized Convex Clustering Optimization and Feature Selection for Mixed Multi-View Data

In mixed multi-view data, multiple sets of diverse features are measured on the same set of samples. By integrating all available data sources, we seek to discover common group structure among the samples that may be hidden in individualistic cluster analyses of a single data view. While several techniques for such integrative clustering have been explored, we propose and develop a convex formalization that enjoys strong empirical performance and inherits the mathematical properties of increasingly popular convex clustering methods. Specifically, our Integrative Generalized Convex Clustering Optimization (iGecco) method employs different convex distances, losses, or divergences for each of the different data views with a joint convex fusion penalty that leads to common groups. Additionally, integrating mixed multi-view data is often challenging when each data source is high-dimensional. To perform feature selection in such scenarios, we develop an adaptive shifted group-lasso penalty that selects features by shrinking them towards their loss-specific centers. Our so-called iGecco+ approach selects features from each data view that are best for determining the groups, often leading to improved integrative clustering. To solve our problem, we develop a new type of generalized multi-block ADMM algorithm using sub-problem approximations that more efficiently fits our model for big data sets. Through a series of numerical experiments and real data examples on text mining and genomics, we show that iGecco+ achieves superior empirical performance for high-dimensional mixed multi-view data.

Integrative Network Fusion: A Multi-Omics Approach in Molecular Profiling

Recent technological advances and international efforts, such as The Cancer Genome Atlas (TCGA), have made available several pan-cancer datasets encompassing multiple omics layers with detailed clinical information in large collection of samples. The need has thus arisen for the development of computational methods aimed at improving cancer subtyping and biomarker identification from multi-modal data. Here we apply the Integrative Network Fusion (INF) pipeline, which combines multiple omics layers exploiting Similarity Network Fusion (SNF) within a machine learning predictive framework. INF includes a feature ranking scheme (rSNF) on SNF-integrated features, used by a classifier over juxtaposed multi-omics features (juXT). In particular, we show instances of INF implementing Random Forest (RF) and linear Support Vector Machine (LSVM) as the classifier, and two baseline RF and LSVM models are also trained on juXT. A compact RF model, called rSNFi, trained on the intersection of top-ranked biomarkers from the two approaches juXT and rSNF is finally derived. All the classifiers are run in a 10x5-fold cross-validation schema to warrant reproducibility, following the guidelines for an unbiased Data Analysis Plan by the US FDA-led initiatives MAQC/SEQC. INF is demonstrated on four classification tasks on three multi-modal TCGA oncogenomics datasets. Gene expression, protein expression and copy number variants are used to predict estrogen receptor status (BRCA-ER, N = 381) and breast invasive carcinoma subtypes (BRCA-subtypes, N = 305), while gene expression, miRNA expression and methylation data is used as predictor layers for acute myeloid leukemia and renal clear cell carcinoma survival (AML-OS, N = 157; KIRC-OS, N = 181). In test, INF achieved similar Matthews Correlation Coefficient (MCC) values and 97% to 83% smaller feature sizes (FS), compared with juXT for BRCA-ER (MCC: 0.83 vs. 0.80; FS: 56 vs. 1801) and BRCA-subtypes (0.84 vs. 0.80; 302 vs. 1801), improving KIRC-OS performance (0.38 vs. 0.31; 111 vs. 2319). INF predictions are generally more accurate in test than one-dimensional omics models, with smaller signatures too, where transcriptomics consistently play the leading role. Overall, the INF framework effectively integrates multiple data levels in oncogenomics classification tasks, improving over the performance of single layers alone and naive juxtaposition, and provides compact signature sizes.

A 17-gene expression-based prognostic signature associated with the prognosis of patients with breast cancer: A STROBE-compliant study

Identification of reliable predictive biomarkers for patients with breast cancer (BC).Univariate Cox proportional hazards regression model was conducted to identify genes correlated with the overall survival (OS) of patients in the TCGA-BRCA cohort. Functional enrichment analysis was conducted to investigate the biological meaning of these survival related genes. Then, patients in TCGA-BCRA were randomly divided into training set and test. Least absolute shrinkage and selection operator (LASSO) penalized Cox regression model was performed and the risk score of BC patients in this model was used to build a prognostic signature. The prognostic performance of the signature was evaluated in the training set, test set, and an independent validation set GSE7390.2519 genes were demonstrated to be significantly associated with the OS of BC patients. Functional annotation of the 2519 genes suggested that these genes were associated with immune response and protein synthesis related gene ontology terms and pathways. 17 genes were identified in the LASSO Cox regression model and used to construct a 17-gene signature. Patients in the 17-gene signature low risk group have better OS and event-free survival compared with those in the 17-gene signature high risk group in the TCGA-BRCA cohort. The prognostic role of the 17-gene signature has been confirmed in the validation cohort. Multivariable Cox proportional hazards regression model suggested the 17-gene signature was an independent prognostic factor in BC.The 17-gene signature we developed could successfully classify patients into high- and low-risk groups, indicating that it might serve as candidate biomarker in BC.

Radiogenomic Analysis of Breast Cancer by Using B-Mode and Vascular US and RNA Sequencing

Background Radiogenomic investigations for breast cancer provide an understanding of tumor heterogeneity and discover image phenotypes of genetic variation. However, there is little research on the correlations between US features of breast cancer and whole-transcriptome profiling. Purpose To explore US phenotypes reflecting genetic alteration relevant to breast cancer treatment and prognosis by comparing US images of tumor with their RNA sequencing results. Materials and Methods From January to October 2016, B-mode and vascular US images in 31 women (mean age, 49 years ± 9 [standard deviation]) with breast cancer were prospectively analyzed. B-mode features included size, shape, echo pattern, orientation, margin, and calcifications. Vascular features were evaluated by using microvascular US and contrast agent-enhanced US: vascular index, vessel morphologic features, distribution, penetrating vessels, enhancement degree, order, margin, internal homogeneity, and perfusion defect. RNA sequencing was conducted with total RNA obtained from a surgical specimen by using next-generation sequencing. US features were compared with gene expression profiles, and ingenuity pathway analysis was used to analyze gene networks, enriched functions, and canonical pathways associated with breast cancer. The P value for differential expression was extracted by using a parametric F test comparing nested linear models. Results Thirteen US features were associated with various patterns of 340 genes (P < .05). Nonparallel orientation at B-mode US was associated with upregulation of TFF1 (log twofold change [log2FC] = 4.0; P < .001), TFF3 (log2FC = 2.5; P < .001), AREG (log2FC = 2.6; P = .005), and AGR3 (log2FC = 2.6; P = .003). Complex vessel morphologic structure was associated with upregulation of FZD8 (log2FC = 2.0; P = .01) and downregulation of IGF1R (log2FC = -2.0; P = .006) and CRIPAK (log2FC = -2.4; P = .01). The top networks with regard to orientation or vessel morphologic structure were associated with cell cycle, death, and proliferation. Conclusion Compared with RNA sequencing, B-mode and vascular US features reflected genomic alterations associated with hormone receptor status, angiogenesis, or prognosis in breast cancer. © RSNA, 2020 Online supplemental material is available for this article.

Extracellular AGR3 regulates breast cancer cells migration via Src signaling

Human anterior gradient proteins AGR2 and AGR3 are overexpressed in a variety of adenocarcinomas and are often secreted in cancer patients' specimens, which suggests a role for AGR proteins in intra and extracellular compartments. Although these proteins exhibit high sequence homology, AGR2 is predominantly described as a pro-oncogene and a potential prognostic biomarker. However, little is known about the function of AGR3. Therefore, the aim of the present study was to investigate the role of AGR3 in breast cancer. The results demonstrated that breast cancer cells secrete AGR3. Furthermore, it was revealed that extracellular AGR3 (eAGR3) regulates tumor cell adhesion and migration. The current study indicated that the pharmacological and genetic perturbation of Src kinase signaling, through treatment with Dasatinib (protein kinase inhibitor) or investigating cells that express a dominant-negative form of Src, significantly abrogated eAGR3-mediated breast cancer cell migration. Therefore, the results indicated that eAGR3 may control tumor cell migration via activation of Src kinases. The results of the present study indicated that eAGR3 may serve as a microenvironmental signaling molecule in tumor-associated processes.

Evaluation of variable selection methods for random forests and omics data sets

Machine learning methods and in particular random forests are promising approaches for prediction based on high dimensional omics data sets. They provide variable importance measures to rank predictors according to their predictive power. If building a prediction model is the main goal of a study, often a minimal set of variables with good prediction performance is selected. However, if the objective is the identification of involved variables to find active networks and pathways, approaches that aim to select all relevant variables should be preferred. We evaluated several variable selection procedures based on simulated data as well as publicly available experimental methylation and gene expression data. Our comparison included the Boruta algorithm, the Vita method, recurrent relative variable importance, a permutation approach and its parametric variant (Altmann) as well as recursive feature elimination (RFE).

 In our simulation studies, Boruta was the most powerful approach, followed closely by the Vita method. Both approaches demonstrated similar stability in variable selection, while Vita was the most robust approach under a pure null model without any predictor variables related to the outcome. In the analysis of the different experimental data sets, Vita demonstrated slightly better stability in variable selection and was less computationally intensive than Boruta.

In conclusion, we recommend the Boruta and Vita approaches for the analysis of high-dimensional data sets. Vita is considerably faster than Boruta and thus more suitable for large data sets, but only Boruta can also be applied in low-dimensional settings.

Anterior Gradient 3 Promotes Breast Cancer Development and Chemotherapy Response

PURPOSE

Anterior gradient 3 (AGR3) belongs to human anterior gradient (AGR) family. The function of AGR3 on cancer remains unknown. This research aimed to investigate if AGR3 had prognostic values in invasive ductal carcinoma (IDC) of breast cancer and could promote tumor progression.

Materials and Methods

AGR3 expression was detected in breast benign lesions, ductal carcinoma in situ and IDC by immunohistochemistry analysis. AGR3's correlations with clinicopathological features and prognosis of IDC patients were analyzed. By cell function experiments, collagen gel droplet-embedded culture drug sensitivity test and cytotoxic analysis, AGR3's impacts on proliferation, invasion ability, and chemotherapeutic drug sensitivity of breast cancer cells were also detected.

RESULTS

AGR3 was up-regulated in luminal subtype of histological grade I-II of IDC patients and positively correlated with high risks of recurrence and distant metastasis. AGR3 high expression could lead to bone or liver metastasis and predict poor prognosis of luminal B. In cell lines, AGR3 could promote proliferation and invasion ability of breast cancer cells which were consistent with clinical analysis. Besides, AGR3 could indicate poor prognosis of breast cancer patients treated with taxane but a favorable prognosis with 5-fluoropyrimidines. And breast cancer cells with AGR3 high expression were resistant to taxane but sensitive to 5-fluoropyrimidines.

CONCLUSION

AGR3 might be a potential prognostic indicator in luminal B subtype of IDC patients of histological grade I-II. And patients with AGR3 high expression should be treated with chemotherapy regimens consisting of 5-fluoropyrimidines but no taxane.

Assessing the efficacy of androgen receptor and Sox10 as independent markers of the triple-negative breast cancer subtype by transcriptome profiling

The Androgen Receptor (AR) has recently garnered a lot of attention as a potential biomarker and therapeutic target in hormone-dependent cancers, including breast cancer. However, several inconsistencies exist within the literature as to which subtypes of breast cancer express AR or whether it can be used to define its own unique subtype. Here, we analyze 1246 invasive breast cancer samples from the Cancer Genome Atlas and show that human breast cancers that have been subtyped based on their HER2, ESR1, or PGR expression contain four clusters of genes that are differentially expressed across all subtypes. We demonstrate that Sox10 is highly expressed in approximately one-third of all HER2/ESR1/PGR-low tumors and is a candidate biomarker of the triple-negative subtype. Although AR expression is acquired in many breast cancer cases, its expression could not define a unique subtype. Despite several reports stating that AR expression is acquired in HER2/ESR1/PGR triple-negative cancers, here we show that a low percentage of these cancers express AR (~20%). In contrast, AR is highly expressed in HER2-positive or ESR1/PGR-positive cancers (> 95%). Although AR expression cannot be used as an independent subtype biomarker, our analysis shows that routine evaluation of AR expression in tumors which express HER2, ESR1 and/or PGR may identify a unique subset of tumors which would benefit from anti-androgen based therapies.

Crystal structure of human anterior gradient protein 3

Oxidative protein folding in the endoplasmic reticulum is catalyzed by the protein disulfide isomerase family of proteins. Of the 20 recognized human family members, the structures of eight have been deposited in the PDB along with domains from six more. Three members of this family, ERp18, anterior gradient protein 2 (AGR2) and anterior gradient protein 3 (AGR3), are single-domain proteins which share sequence similarity. While ERp18 has a canonical active-site motif and is involved in native disulfide-bond formation, AGR2 and AGR3 lack elements of the active-site motif found in other family members and may both interact with mucins. In order to better define its function, the structure of AGR3 is required. Here, the recombinant expression, purification, crystallization and crystal structure of human AGR3 are described.

Enhancer-Mediated Oncogenic Function of the Menin Tumor Suppressor in Breast Cancer

While the multiple endocrine neoplasia type 1 (MEN1) gene functions as a tumor suppressor in a variety of cancer types, we explored its oncogenic role in breast tumorigenesis. The MEN1 gene product menin is involved in H3K4 trimethylation and co-activates transcription. We integrated ChIP-seq and RNA-seq data to identify menin target genes. Our analysis revealed that menin-dependent target gene promoters display looping to distal enhancers that are bound by menin, FOXA1 and GATA3. In this fashion, MEN1 co-regulates a proliferative breast cancer-specific gene expression program in ER⁺ cells. In primary mammary cells, MEN1 exerts an anti-proliferative function by regulating a distinct expression signature. Our findings clarify the cell-type-specific functions of MEN1 and inform the development of menin-directed treatments for breast cancer.

Expression of protein disulfide isomerase family members correlates with tumor progression and patient survival in ovarian cancer

Objective

Protein disulfide isomerase (PDI) is an oxidoreductase that is overexpressed in several cancers. PDI family members (PDIs) play a role in various diseases including cancer. Select PDIs were reported as useful markers in other cancers but their expression in ovarian cancer has not been thoroughly assessed. We sought to evaluate the expression of PDI, PDIA6, PDIR, ERp57, ERp72 and AGR3 in ovarian cancer patient samples and examine their prognostic significance.

Methods

TMA samples from 415 tissues collected from three cancer centers (UM, USC, and KCCRI) were used to assess the expression levels of PDI family proteins using IHC.

Results

We observed significant increases in PDI (p = 9.16E-36), PDIA6 (p = 5.51E-33), PDIR (p = 1.81E-12), ERp57 (p = 9.13E-07), ERp72 (p = 3.65E-22), and AGR3 (p = 4.56E-24) expression in ovarian cancers compared to normal tissues. Expression of PDI family members also increases during disease progression (p <0.001). All PDI family members are overexpressed in serous ovarian cancer (p<0.001). However, PDI, PDIA6, PDIR, ERp72 and AGR3 are more significantly overexpressed (p<0.001) than ERp57 (p<0.05) in clear cell ovarian carcinoma. Importantly, overexpression of PDI family members is associated with poor survival in ovarian cancer (p = 0.045 for PDI, p = 0.047 for PDIR, p = 0.037 for ERp57, p = 0.046 for ERp72, p = 0.040 for AGR3) with the exception of PDIA6 (p = 0.381).

Conclusions

Our findings demonstrate that select PDI family members (PDI, PDIR, ERp72, ERp57 and AGR3) are potential prognostic markers for ovarian cancer.

Prognostic genes of breast cancer revealed by gene co-expression network analysis

The aim of the present study was to identify genes that may serve as markers for breast cancer prognosis by constructing a gene co-expression network and mining modules associated with survival. Two gene expression datasets of breast cancer were downloaded from ArrayExpress and genes from these datasets with a coefficient of variation >0.5 were selected and underwent functional enrichment analysis with the Database for Annotation, Visualization and Integration Discovery. Gene co-expression networks were constructed with the WGCNA package in R. Modules were identified from the network via cluster analysis. Cox regression was conducted to analyze survival rates. A total of 2,669 genes were selected, and functional enrichment analysis of them revealed that they were mainly associated with the immune response, cell proliferation, cell differentiation and cell adhesion. Seven modules were identified from the gene co-expression network, one of which was found to be significantly associated with patient survival time. Expression status of 144 genes from this module was used to cluster patient samples into two groups, with a significant difference in survival time revealed between these groups. These genes were involved in the cell cycle and tumor protein p53 signaling pathway. The top 10 hub genes were identified in the module. The findings of the present study could advance the understanding of the molecular pathogenesis of breast cancer.

Screening and identification of potential biomarkers in triple-negative breast cancer by integrated analysis

Triple-negative breast cancer (TNBC) has attracted great attention due to its unique biology, poor prognosis, and aggressiveness. TNBC patients are more likely to suffer from metastasis. We screened and identified the TNBC-specific genes as potential biomarkers. A total of 167 breast cancer samples (45 TNBC and 122 non-TNBC) were used in the integrated analysis. Gene expression microarrays were used to screen the differentially expressed genes. We identified 65 core DEGs. According to the GO and KEGG analysis, the gene function enrichment in TNBC was revealed, such as basal cell carcinoma, prostate cancer, oocyte meiosis and choline metabolism in cancer pathways. Moreover, the PPI network reconstruction would benefit the screening of hubs. A RFS analysis of TNBC-specific genes was also conducted. RT-PCR was used to validate the expression pattern of hubs in TNBC. Finally, nine genes were identified and all of them were novel, specific and higher dysregulation expressed genes in TNBC. Such that, these genes will serve as potential biomarkers in TNBC and benefit further research in TNBC.

Identification and prognostic value of anterior gradient protein 2 expression in breast cancer based on tissue microarray

Breast cancer has attracted substantial attention as one of the major cancers causing death in women. It is crucial to find potential biomarkers of prognostic value in breast cancer. In this study, the expression pattern of anterior gradient protein 2 in breast cancer was identified based on the main molecular subgroups. Through analysis of 69 samples from the Gene Expression Omnibus database, we found that anterior gradient protein 2 expression was significantly higher in non-triple-negative breast cancer tissues compared with normal tissues and triple-negative breast cancer tissues (p < 0.05). The data from a total of 622 patients from The Cancer Genome Atlas were analysed. The data from The Cancer Genome Atlas and results from quantitative reverse transcription polymerase chain reaction also verified the anterior gradient protein 2 expression pattern. Furthermore, we performed immunohistochemical analysis. The quantification results revealed that anterior gradient protein 2 is highly expressed in non-triple-negative breast cancer (grade 3 excluded) and grade 1 + 2 (triple-negative breast cancer excluded) tumours compared with normal tissues. Anterior gradient protein 2 was significantly highly expressed in non-triple-negative breast cancer (grade 3 excluded) and non-triple-negative breast cancer tissues compared with triple-negative breast cancer tissues (p < 0.01). In addition, anterior gradient protein 2 was significantly highly expressed in grade 1 + 2 (triple-negative breast cancer excluded) and grade 1 + 2 tissues compared with grade 3 tissues (p < 0.05). Analysis by Fisher's exact test revealed that anterior gradient protein 2 expression was significantly associated with histologic type, histological grade, oestrogen status and progesterone status. Univariate analysis of clinicopathological variables showed that anterior gradient protein 2 expression, tumour size and lymph node status were significantly correlated with overall survival in patients with grade 1 and 2 tumours. Cox multivariate analysis revealed anterior gradient protein 2 as a putative independent indicator of unfavourable outcomes (p = 0.031). All these data clearly showed that anterior gradient protein 2 is highly expressed in breast cancer and can be regarded as a putative biomarker for breast cancer prognosis.

MINT: a multivariate integrative method to identify reproducible molecular signatures across independent experiments and platforms

Background

Molecular signatures identified from high-throughput transcriptomic studies often have poor reliability and fail to reproduce across studies. One solution is to combine independent studies into a single integrative analysis, additionally increasing sample size. However, the different protocols and technological platforms across transcriptomic studies produce unwanted systematic variation that strongly confounds the integrative analysis results. When studies aim to discriminate an outcome of interest, the common approach is a sequential two-step procedure; unwanted systematic variation removal techniques are applied prior to classification methods.

Results

To limit the risk of overfitting and over-optimistic results of a two-step procedure, we developed a novel multivariate integration method, MINT, that simultaneously accounts for unwanted systematic variation and identifies predictive gene signatures with greater reproducibility and accuracy. In two biological examples on the classification of three human cell types and four subtypes of breast cancer, we combined high-dimensional microarray and RNA-seq data sets and MINT identified highly reproducible and relevant gene signatures predictive of a given phenotype. MINT led to superior classification and prediction accuracy compared to the existing sequential two-step procedures.

Conclusions

MINT is a powerful approach and the first of its kind to solve the integrative classification framework in a single step by combining multiple independent studies. MINT is computationally fast as part of the mixOmics R CRAN package, available at http://www.mixOmics.org/mixMINT/and http://cran.r-project.org/web/packages/mixOmics/.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-017-1553-8) contains supplementary material, which is available to authorized users.