Integrative analysis of survival-associated gene sets in breast cancer

Using hidden Markov model to predict recurrence of breast cancer based on sequential patterns in gene expression profiles

A new approach is presented to predict breast cancer recurrence through gene expression profiles using hidden Markov models (HMM). In this regard, 322 genes were selected from 44 published gene lists related to breast cancer prognosis. Afterwards, using gene set enrichment analysis, 922 gene sets were found from subsets of genes with the same biological meaning. In order to extract the sequential patterns from gene expression data, we ranked the gene sets using appropriate criteria and used HMM in which the ranked gene sets considered as observation sequences and hidden states represented priority of gene sets for discriminating between expression profiles. In this experiment, seven publicly available microarray datasets, including 1271 breast tumor samples, were used to classify cancer patients into two groups according to risk of recurrence. Our experiments indicated the greater performance and more robustness of the proposed model compared with other widely used classification methods.

Detecting Cancer Survival Related Gene Markers Based on Rectified Factor Network

Detecting gene sets that serve as biomarkers for differentiating patient survival groups may help diagnose diseases robustly and develop multi-gene targeted therapies. However, due to the exponential growth of search space imposed by gene combinations, the performance of existing methods is still far from satisfactory. In this study, we developed a new method called BISG (BIclustering based Survival-related Gene sets detection) based on a rectified factor network (RFN) model, which allows efficiently biclustering gene subsets. By correlating genes in each significant bicluster with patient survival outcomes using a log-rank test and multi-sampling strategy, multiple survival-related gene sets can be detected. We applied BISG on three different cancer types, and the resulting gene sets were tested as biomarkers for survival analyses. Secondly, we systematically analyzed 12 different cancer datasets. Our analysis shows that the genes in all the survival-related gene sets are mainly from five gene families: microRNA protein coding host genes, zinc fingers C2H2-type, solute carriers, CD (cluster of differentiation) molecules, and ankyrin repeat domain containing genes. Moreover, we found that they are mainly enriched in heme metabolism, apoptosis, hypoxia and inflammatory response-related pathways. We compared BISG with two other methods, GSAS and IPSOV. Results show that BISG can better differentiate patient survival groups in different datasets. The identified biomarkers suggested by our study provide useful hypotheses for further investigation. BISG is publicly available with open source at https://github.com/LingtaoSu/BISG.

Prediction of survival risks with adjusted gene expression through risk-gene networks

MOTIVATION

Network-based analysis of biomedical data has been extensively studied over the last decades. As a successful application, gene networks have been used to illustrate interactions among genes and explain the associated phenotypes. However, the gene network approaches have not been actively applied for survival analysis, which is one of the main interests of biomedical research. In addition, a few previous studies using gene networks for survival analysis construct networks mainly from prior knowledge, such as pathways, regulations and gene sets, while the performance considerably depends on the selection of prior knowledge.

RESULTS

In this paper, we propose a data-driven construction method for survival risk-gene networks as well as a survival risk prediction method using the network structure. The proposed method constructs risk-gene networks with survival-associated genes using penalized regression. Then, gene expression indices are hierarchically adjusted through the networks to reduce the variance intrinsic in datasets. By illustrating risk-gene structure, the proposed method is expected to provide an intuition for the relationship between genes and survival risks. The risk-gene network is applied to a low grade glioma dataset, and produces a hypothesis of the relationship between genetic biomarkers of low and high grade glioma. Moreover, with multiple datasets, we demonstrate that the proposed method shows superior prediction performance compared to other conventional methods.

AVAILABILITY AND IMPLEMENTATION

The R package of risk-gene networks is freely available in the web at http://cdal.korea.ac.kr/NetDA/.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Cytokine sensitivity screening highlights BMP4 pathway signaling as a therapeutic opportunity in ER+ breast cancer

Despite the success of approved systemic therapies for estrogen receptor α (ER)-positive breast cancer, drug resistance remains common. We hypothesized that secreted factors from the human tumor microenvironment could modulate drug resistance. We previously screened a library of 297 recombinant-secreted microenvironmental proteins for the ability to confer resistance to the anti-estrogen fulvestrant in 2 ER⁺ breast cancer cell lines. Herein, we considered whether factors that enhanced drug sensitivity could be repurposed as therapeutics and provide leads for drug development. Screening data revealed bone morphogenic protein (BMP)4 as a factor that inhibited cell growth and synergized with approved anti-estrogens and cyclin-dependent kinase 4/6 inhibitors (CDK4/6_i). BMP4-mediated growth inhibition was dependent on type I receptor activin receptor-like kinase (ALK)3-dependent phosphorylation (P) of mothers against decapentaplegic homolog (SMAD/P-SMAD)1 and 5, which could be reversed by BMP receptor inhibitors and ALK3 knockdown. The primary effect of BMP4 on cell fate was cell-cycle arrest, in which RNA sequencing, immunoblot analysis, and RNA interference revealed to be dependent on p21^WAF1/Cip1 upregulation. BMP4 also enhanced sensitivity to approved inhibitors of mammalian target of rapamycin complex 1 and CDK4/6 via ALK3-mediated P-SMAD1/5 and p21 upregulation in anti-estrogen-resistant cells. Patients bearing primary ER⁺ breast tumors, exhibiting a transcriptomic signature of BMP4 signaling, had improved disease outcome following adjuvant treatment with anti-estrogen therapy, independently of age, tumor grade, and tumor stage. Furthermore, a transcriptomic signature of BMP4 signaling was predictive of an improved biologic response to the CDK4/6_i palbociclib, in combination with an aromatase inhibitor in primary tumors. These findings highlight BMP4 and its downstream pathway activation as a therapeutic opportunity in ER⁺ breast cancer.-Shee, K., Jiang, A., Varn, F. S., Liu, S., Traphagen, N. A., Owens, P., Ma, C. X., Hoog, J., Cheng, C., Golub, T. R., Straussman, R., Miller, T. W. Cytokine sensitivity screening highlights BMP4 pathway signaling as a therapeutic opportunity in ER⁺ breast cancer.

miRNA in a multiomic context for diagnosis, treatment monitoring and personalized management of metastatic breast cancer

Metastatic breast cancer is characterized by aggressive spreading to distant organs. Despite huge multilevel research, there are still several important challenges that have to be clarified in the management of this disease. Therefore, recent investigations have implemented a modern, multiomic approach with the aim of identifying specific biomarkers for not only early detection but also to predict treatment responses and metastatic spread. Specific attention is paid to short miRNAs, which regulate gene expression at the post-transcriptional level. Aberrant miRNA expression could initiate cancer development, cell proliferation, invasion, migration, metastatic spread or drug resistance. An miRNA signature is, therefore, believed to be a promising biomarker and prediction tool that could be utilized in all phases of carcinogenesis. This article offers comprehensive information about miRNA profiles useful for diagnostic and treatment purposes that may sufficiently advance breast cancer management and improve individual outcomes in the near future.

MOBCdb: a comprehensive database integrating multi-omics data on breast cancer for precision medicine

BACKGROUND

Breast cancer is one of the most frequently diagnosed cancers among women worldwide, characterized by diverse biological heterogeneity. It is well known that complex and combined gene regulation of multi-omics is involved in the occurrence and development of breast cancer.

RESULTS

In this paper, we present the Multi-Omics Breast Cancer Database (MOBCdb), a simple and easily accessible repository that integrates genomic, transcriptomic, epigenomic, clinical, and drug response data of different subtypes of breast cancer. MOBCdb allows users to retrieve simple nucleotide variation (SNV), gene expression, microRNA expression, DNA methylation, and specific drug response data by various search fashions. The genome-wide browser /navigation facility in MOBCdb provides an interface for visualizing multi-omics data of multi-samples simultaneously. Furthermore, the survival module provides survival analysis for all or some of the samples by using data of three omics. The approved public drugs with genetic variations on breast cancer are also included in MOBCdb.

CONCLUSION

In summary, MOBCdb provides users a unique web interface to the integrated multi-omics data of different subtypes of breast cancer, which enables the users to identify potential novel biomarkers for precision medicine.

Therapeutically targeting tumor microenvironment-mediated drug resistance in estrogen receptor-positive breast cancer

Tumor microenvironment (TME) cytokine screening revealed FGF2 as a clinically relevant mechanism of resistance to anti-estrogens, mTORC1 inhibition, and PI3K inhibition in ER+ breast cancer. Shee et al. highlight an underdeveloped aspect of precision oncology: treating patients according to their TME constitution.

Drug resistance to approved systemic therapies in estrogen receptor–positive (ER+) breast cancer remains common. We hypothesized that factors present in the human tumor microenvironment (TME) drive drug resistance. Screening of a library of recombinant secreted microenvironmental proteins revealed fibroblast growth factor 2 (FGF2) as a potent mediator of resistance to anti-estrogens, mTORC1 inhibition, and phosphatidylinositol 3-kinase inhibition in ER+ breast cancer. Phosphoproteomic analyses identified ERK1/2 as a major output of FGF2 signaling via FGF receptors (FGFRs), with consequent up-regulation of Cyclin D1 and down-regulation of Bim as mediators of drug resistance. FGF2-driven drug resistance in anti-estrogen–sensitive and –resistant models, including patient-derived xenografts, was reverted by neutralizing FGF2 or FGFRs. A transcriptomic signature of FGF2 signaling in primary tumors predicted shorter recurrence-free survival independently of age, grade, stage, and FGFR amplification status. These findings delineate FGF2 signaling as a ligand-based drug resistance mechanism and highlights an underdeveloped aspect of precision oncology: characterizing and treating patients according to their TME constitution.

Comparison of multiple transcriptomes exposes unified and divergent features of quiescent and activated skeletal muscle stem cells

BACKGROUND

Skeletal muscle satellite (stem) cells are quiescent in adult mice and can undergo multiple rounds of proliferation and self-renewal following muscle injury. Several labs have profiled transcripts of myogenic cells during the developmental and adult myogenesis with the aim of identifying quiescent markers. Here, we focused on the quiescent cell state and generated new transcriptome profiles that include subfractionations of adult satellite cell populations, and an artificially induced prenatal quiescent state, to identify core signatures for quiescent and proliferating.

METHODS

Comparison of available data offered challenges related to the inherent diversity of datasets and biological conditions. We developed a standardized workflow to homogenize the normalization, filtering, and quality control steps for the analysis of gene expression profiles allowing the identification up- and down-regulated genes and the subsequent gene set enrichment analysis. To share the analytical pipeline of this work, we developed Sherpa, an interactive Shiny server that allows multi-scale comparisons for extraction of desired gene sets from the analyzed datasets. This tool is adaptable to cell populations in other contexts and tissues.

RESULTS

A multi-scale analysis comprising eight datasets of quiescent satellite cells had 207 and 542 genes commonly up- and down-regulated, respectively. Shared up-regulated gene sets include an over-representation of the TNFα pathway via NFKβ signaling, Il6-Jak-Stat3 signaling, and the apical surface processes, while shared down-regulated gene sets exhibited an over-representation of Myc and E2F targets and genes associated to the G2M checkpoint and oxidative phosphorylation. However, virtually all datasets contained genes that are associated with activation or cell cycle entry, such as the immediate early stress response genes Fos and Jun. An empirical examination of fixed and isolated satellite cells showed that these and other genes were absent in vivo, but activated during procedural isolation of cells.

CONCLUSIONS

Through the systematic comparison and individual analysis of diverse transcriptomic profiles, we identified genes that were consistently differentially expressed among the different datasets and shared underlying biological processes key to the quiescent cell state. Our findings provide impetus to define and distinguish transcripts associated with true in vivo quiescence from those that are first responding genes due to disruption of the stem cell niche.

Aberrant expression of epithelial leucine-rich repeat containing G protein-coupled receptor 5-positive cells in the eutopic endometrium in endometriosis and implications in deep-infiltrating endometriosis

OBJECTIVE

To characterize leucine-rich repeat containing G protein-coupled receptor 5-positive (LGR5⁺) cells from the endometrium of women with endometriosis.

DESIGN

Prospective experimental study.

SETTING

University hospital/fertility clinic.

PATIENT(S)

Twenty-seven women with endometriosis who underwent surgery and 12 healthy egg donors, together comprising 39 endometrial samples.

INTERVENTION(S)

Obtaining of uterine aspirates by using a Cornier Pipelle.

MAIN OUTCOMES MEASURE(S)

Immunofluorescence in formalin-fixed paraffin-embedded tissue from mice and healthy and pathologic human endometrium using antibodies against LGR5, E-cadherin, and cytokeratin, and epithelial and stromal LGR5⁺ cells isolated from healthy and pathologic human eutopic endometrium by fluorescence-activated cell sorting and transcriptomic characterization by RNA high sequencing.

RESULT(S)

Immunofluorescence showed that LGR5⁺ cells colocalized with epithelial markers in the stroma of the endometrium only in endometriotic patients. The results from RNA high sequencing of LGR5⁺ cells from epithelium and stroma did not show any statistically significant differences between them. The LGR5⁺ versus LGR5^- cells in pathologic endometrium showed 394 differentially expressed genes. The LGR5⁺ cells in deep-infiltrating endometriosis expressed inflammatory markers not present in the other types of the disease.

CONCLUSION(S)

Our results revealed the presence of aberrantly located LGR5⁺ cells coexpressing epithelial markers in the stromal compartment of women with endometriosis. These cells have a statistically significantly different expression profile in deep-infiltrating endometriosis in comparison with other types of endometriosis, independent of the menstrual cycle phase. Further studies are needed to elucidate their role and influence in reproductive outcomes.

A novel pathway-based distance score enhances assessment of disease heterogeneity in gene expression

BACKGROUND

Distance based unsupervised clustering of gene expression data is commonly used to identify heterogeneity in biologic samples. However, high noise levels in gene expression data and relatively high correlation between genes are often encountered, so traditional distances such as Euclidean distance may not be effective at discriminating the biological differences between samples. An alternative method to examine disease phenotypes is to use pre-defined biological pathways. These pathways have been shown to be perturbed in different ways in different subjects who have similar clinical features. We hypothesize that differences in the expressions of genes in a given pathway are more predictive of differences in biological differences compared to standard approaches and if integrated into clustering analysis will enhance the robustness and accuracy of the clustering method. To examine this hypothesis, we developed a novel computational method to assess the biological differences between samples using gene expression data by assuming that ontologically defined biological pathways in biologically similar samples have similar behavior.

RESULTS

Pre-defined biological pathways were downloaded and genes in each pathway were used to cluster samples using the Gaussian mixture model. The clustering results across different pathways were then summarized to calculate the pathway-based distance score between samples. This method was applied to both simulated and real data sets and compared to the traditional Euclidean distance and another pathway-based clustering method, Pathifier. The results show that the pathway-based distance score performs significantly better than the Euclidean distance, especially when the heterogeneity is low and genes in the same pathways are correlated. Compared to Pathifier, we demonstrated that our approach achieves higher accuracy and robustness for small pathways. When the pathway size is large, by downsampling the pathways into smaller pathways, our approach was able to achieve comparable performance.

CONCLUSIONS

We have developed a novel distance score that represents the biological differences between samples using gene expression data and pre-defined biological pathway information. Application of this distance score results in more accurate, robust, and biologically meaningful clustering results in both simulated data and real data when compared to traditional methods. It also has comparable or better performance compared to Pathifier.

Contextual Refinement of Regulatory Targets Reveals Effects on Breast Cancer Prognosis of the Regulome

Gene expression regulators, such as transcription factors (TFs) and microRNAs (miRNAs), have varying regulatory targets based on the tissue and physiological state (context) within which they are expressed. While the emergence of regulator-characterizing experiments has inferred the target genes of many regulators across many contexts, methods for transferring regulator target genes across contexts are lacking. Further, regulator target gene lists frequently are not curated or have permissive inclusion criteria, impairing their use. Here, we present a method called iterative Contextual Transcriptional Activity Inference of Regulators (icTAIR) to resolve these issues. icTAIR takes a regulator’s previously-identified target gene list and combines it with gene expression data from a context, quantifying that regulator’s activity for that context. It then calculates the correlation between each listed target gene’s expression and the quantitative score of regulatory activity, removes the uncorrelated genes from the list, and iterates the process until it derives a stable list of refined target genes. To validate and demonstrate icTAIR’s power, we use it to refine the MSigDB c3 database of TF, miRNA and unclassified motif target gene lists for breast cancer. We then use its output for survival analysis with clinicopathological multivariable adjustment in 7 independent breast cancer datasets covering 3,430 patients. We uncover many novel prognostic regulators that were obscured prior to refinement, in particular NFY, and offer a detailed look at the composition and relationships among the breast cancer prognostic regulome. We anticipate icTAIR will be of general use in contextually refining regulator target genes for discoveries across many contexts. The icTAIR algorithm can be downloaded from https://github.com/icTAIR.

Gene expression regulators, such as transcription factors and microRNAs, are critical actors in cellular physiology and pathophysiology and act by modulating the expression levels of sets of target genes. Given their significance, numerous experiments have sought to characterize the specific target genes of specific regulators, which in turn has led to regulator target gene list databases. Unfortunately, these lists are plagued by poor curation and validation. Further, all lists suffer from the fundamental issue that regulator targets vary across tissue type and physiological state, or “context”, making them poor for conducting downstream, context-specific analyses. To address this issue, here we present a method called icTAIR that contextually-refines regulator target gene lists. To demonstrate its value, we use icTAIR to take the largest-available database of regulator target gene lists, refine it for the breast cancer context, and use both the pre-refined and refined lists for downstream survival analyses in over 3,400 tumors. We find that icTAIR improves the statistical power of the analyses by multiple orders of magnitude. This in turn lets us map the relational network of breast cancer regulators and identify regulators with prognostic effects even after clinicopathological adjustment. We anticipate icTAIR will be broadly useful in regulator studies.

Adaptive immunity programmes in breast cancer

The role of the immune system in shaping cancer development and patient prognosis has recently become an area of intense focus in industry and academia. Harnessing the adaptive arm of the immune system for tumour eradication has shown great promise in a variety of tumour types. Differences between tissues, however, necessitate a greater understanding of the adaptive immunity programmes that are active within each tumour type. In breast cancer, adaptive immune programmes play diverse roles depending on the cellular infiltration found in each tumour. Cytotoxic T lymphocytes and T helper type 1 cells can induce tumour eradication, whereas regulatory T cells and T helper type 2 cells are known to be involved in tumour-promoting immunosuppressive responses. Complicating these matters, heterogeneous expression of hormone receptors and growth factors in different tumours leads to disparate, patient-specific adaptive immune responses. Despite this non-conformity in adaptive immune behaviours, encouraging basic and clinical results have been observed that suggest a role for immunotherapeutic approaches in breast cancer. Here, we review the literature pertaining to the adaptive immune response in breast cancer, summarize the primary findings relating to the breast tumour's biology, and discuss potential clinical immunotherapies.

A network medicine approach to build a comprehensive atlas for the prognosis of human cancer

The Cancer Genome Atlas project has generated multi-dimensional and highly integrated genomic data from a large number of patient samples with detailed clinical records across many cancer types, but it remains unclear how to best integrate the massive amount of genomic data into clinical practice. We report here our methodology to build a multi-dimensional subnetwork atlas for cancer prognosis to better investigate the potential impact of multiple genetic and epigenetic (gene expression, copy number variation, microRNA expression and DNA methylation) changes on the molecular states of networks that in turn affects complex cancer survivorship. We uncover an average of 38 novel subnetworks in the protein-protein interaction network that correlate with prognosis across four prominent cancer types. The clinical utility of these subnetwork biomarkers was further evaluated by prognostic impact evaluation, functional enrichment analysis, drug target annotation, tumor stratification and independent validation. Some pathways including the dynactin, cohesion and pyruvate dehydrogenase-related subnetworks are identified as promising new targets for therapy in specific cancer types. In conclusion, this integrative analysis of existing protein interactome and cancer genomics data allows us to systematically dissect the molecular mechanisms that underlie unexpected outcomes for cancer, which could be used to better understand and predict clinical outcomes, optimize treatment and to provide new opportunities for developing therapeutics related to the subnetworks identified.

Integrative Genomic Analyses Yield Cell-Cycle Regulatory Programs with Prognostic Value

Liposarcoma is the second most common form of sarcoma, which has been categorized into four molecular subtypes, which are associated with differential prognosis of patients. However, the transcriptional regulatory programs associated with distinct histologic and molecular subtypes of liposarcoma have not been investigated. This study uses integrative analyses to systematically define the transcriptional regulatory programs associated with liposarcoma. Likewise, computational methods are used to identify regulatory programs associated with different liposarcoma subtypes, as well as programs that are predictive of prognosis. Further analysis of curated gene sets was used to identify prognostic gene signatures. The integration of data from a variety of sources, including gene expression profiles, transcription factor-binding data from ChIP-Seq experiments, curated gene sets, and clinical information of patients, indicated discrete regulatory programs (e.g., controlled by E2F1 and E2F4), with significantly different regulatory activity in one or multiple subtypes of liposarcoma with respect to normal adipose tissue. These programs were also shown to be prognostic, wherein liposarcoma patients with higher E2F4 or E2F1 activity associated with unfavorable prognosis. A total of 259 gene sets were significantly associated with patient survival in liposarcoma, among which > 50% are involved in cell cycle and proliferation.

IMPLICATIONS

These integrative analyses provide a general framework that can be applied to investigate the mechanism and predict prognosis of different cancer types.

Systematic analysis of hematopoietic gene expression profiles for prognostic prediction in acute myeloid leukemia

Acute myeloid leukemia (AML) is a hematopoietic disorder initiated by the leukemogenic transformation of myeloid cells into leukemia stem cells (LSCs). Preexisting gene expression programs in LSCs can be used to assess their transcriptional similarity to hematopoietic cell types. While this relationship has previously been examined on a small scale, an analysis that systematically investigates this relationship throughout the hematopoietic hierarchy has yet to be implemented. We developed an integrative approach to assess the similarity between AML patient tumor profiles and a collection of 232 murine hematopoietic gene expression profiles compiled by the Immunological Genome Project. The resulting lineage similarity scores (LSS) were correlated with patient survival to assess the relationship between hematopoietic similarity and patient prognosis. This analysis demonstrated that patient tumor similarity to immature hematopoietic cell types correlated with poor survival. As a proof of concept, we highlighted one cell type identified by our analysis, the short-term reconstituting stem cell, whose LSSs were significantly correlated with patient prognosis across multiple datasets, and showed distinct patterns in patients stratified by traditional clinical variables. Finally, we validated our use of murine profiles by demonstrating similar results when applying our method to human profiles.