Discovery and validation of breast cancer subtypes

Exploring Biomarkers in Breast Cancer: Hallmarks of Diagnosis, Treatment, and Follow-Up in Clinical Practice

Breast cancer is a prevalent malignancy in the present day, particularly affecting women as one of the most common forms of cancer. A significant portion of patients initially present with localized disease, for which curative treatments are pursued. Conversely, another substantial segment is diagnosed with metastatic disease, which has a worse prognosis. Recent years have witnessed a profound transformation in the prognosis for this latter group, primarily due to the discovery of various biomarkers and the emergence of targeted therapies. These biomarkers, encompassing serological, histological, and genetic indicators, have demonstrated their value across multiple aspects of breast cancer management. They play crucial roles in initial diagnosis, aiding in the detection of relapses during follow-up, guiding the application of targeted treatments, and offering valuable insights for prognostic stratification, especially for highly aggressive tumor types. Molecular markers have now become the keystone of metastatic breast cancer diagnosis, given the diverse array of chemotherapy options and treatment modalities available. These markers signify a transformative shift in the arsenal of therapeutic options against breast cancer. Their diagnostic precision enables the categorization of tumors with elevated risks of recurrence, increased aggressiveness, and heightened mortality. Furthermore, the existence of therapies tailored to target specific molecular anomalies triggers a cascade of changes in tumor behavior. Therefore, the primary objective of this article is to offer a comprehensive review of the clinical, diagnostic, prognostic, and therapeutic utility of the principal biomarkers currently in use, as well as of their clinical impact on metastatic breast cancer. In doing so, our goal is to contribute to a more profound comprehension of this complex disease and, ultimately, to enhance patient outcomes through more precise and effective treatment strategies.

Clinical and Translational Applications of Serological and Histopathological Biomarkers in Metastatic Breast Cancer: A Comprehensive Review

Breast cancer is one of the most common malignancies worldwide and the most common form of cancer in women. A large proportion of patients begin with localized disease and undergo treatment with curative intent, while another large proportion of patients debuts with disseminated metastatic disease. In the last subgroup of patients, the prognosis in recent years has changed radically, given the existence of different targeted therapies thanks to the discovery of different biomarkers. Serological, histological, and genetic biomarkers have demonstrated their usefulness in the initial diagnosis, in the follow-up to detect relapses, to guide targeted treatment, and to stratify the prognosis of the most aggressive tumors in those with breast cancer. Molecular markers are currently the basis for the diagnosis of metastatic disease, given the wide variety of chemotherapy regions and existing therapies. These markers have been a real revolution in the therapeutic arsenal for breast cancer, and their diagnostic validity allows the classification of tumors with higher rates of relapse, aggressiveness, and mortality. In this sense, the existence of therapies targeting different molecular alterations causes a series of changes in tumor biology that can be assessed throughout the course of the disease to provide information on the underlying pathophysiology of metastatic disease, which allows us to broaden our knowledge of the different mechanisms of tissue invasion. Therefore, the aim of the present article is to review the clinical, diagnostic, predictive, prognostic utility and limitations of the main biomarkers available and under development in metastatic breast cancer.

Position of Circulating Tumor Cells in the Clinical Routine in Prostate Cancer and Breast Cancer Patients

Simple Summary

Many different therapies are applied to fight tumor disease. Blood-based biosources, like circulating tumor cells (CTCs), offer the opportunity to monitor the healing progression and the real-time response to the therapy. In this review, we analyze the outcomes of the clinical trials and scientific studies of prostate and breast cancer performed in the decade between April 2010 and April 2020. Additionally, we describe the abstracts from the 4th Advances in Circulating Tumor Cells (ACTC) meeting in 2019. We discuss the potential therapeutic opportunities related to the CTCs and the challenges ahead in the routine treatment of cancer.

Abstract

Prostate cancer and breast cancer are the most common cancers worldwide. Anti-tumor therapies are long and exhaustive for the patients. The real-time monitoring of the healing progression could be a useful tool to evaluate therapeutic response. Blood-based biosources like circulating tumor cells (CTCs) may offer this opportunity. Application of CTCs for the clinical diagnostics could improve the sequenced screening, provide additional valuable information of tumor dynamics, and help personalized management for the patients. In the past decade, CTCs as liquid biopsy (LB) has received tremendous attention. Many different isolation and characterization platforms are developed but the clinical validation is still missing. In this review, we focus on the clinical trials of circulating tumor cells that have the potential to monitor and stratify patients and lead to implementation into clinical practice.

CONFIGURE: A pipeline for identifying context specific regulatory modules from gene expression data and its application to breast cancer

BACKGROUND

Gene expression data is widely used for identifying subtypes of diseases such as cancer. Differentially expressed gene analysis and gene set enrichment analysis are widely used for identifying biological mechanisms at the gene level and gene set level, respectively. However, the results of differentially expressed gene analysis are difficult to interpret and gene set enrichment analysis does not consider the interactions among genes in a gene set.

RESULTS

We present CONFIGURE, a pipeline that identifies context specific regulatory modules from gene expression data. First, CONFIGURE takes gene expression data and context label information as inputs and constructs regulatory modules. Then, CONFIGURE makes a regulatory module enrichment score (RMES) matrix of enrichment scores of the regulatory modules on samples using the single-sample GSEA method. CONFIGURE calculates the importance scores of the regulatory modules on each context to rank the regulatory modules. We evaluated CONFIGURE on the Cancer Genome Atlas (TCGA) breast cancer RNA-seq dataset to determine whether it can produce biologically meaningful regulatory modules for breast cancer subtypes. We first evaluated whether RMESs are useful for differentiating breast cancer subtypes using a multi-class classifier and one-vs-rest binary SVM classifiers. The multi-class and one-vs-rest binary classifiers were trained using the RMESs as features and outperformed baseline classifiers. Furthermore, we conducted literature surveys on the basal-like type specific regulatory modules obtained by CONFIGURE and showed that highly ranked modules were associated with the phenotypes of basal-like type breast cancers.

CONCLUSIONS

We showed that enrichment scores of regulatory modules are useful for differentiating breast cancer subtypes and validated the basal-like type specific regulatory modules by literature surveys. In doing so, we found regulatory module candidates that have not been reported in previous literature. This demonstrates that CONFIGURE can be used to predict novel regulatory markers which can be validated by downstream wet lab experiments. We validated CONFIGURE on the breast cancer RNA-seq dataset in this work but CONFIGURE can be applied to any gene expression dataset containing context information.

Liquid biopsy in breast cancer: A comprehensive review

Breast cancer is the most common cancer among women worldwide. Due to its complexity in nature, effective breast cancer treatment can encounter many challenges. Traditional methods of cancer detection such as tissue biopsy are not comprehensive enough to capture the entire genomic landscape of breast tumors. However, with the introduction of novel techniques, the application of liquid biopsy has been enhanced, enabling the improvement of various aspects of breast cancer management including early diagnosis and screening, prediction of prognosis, early detection of relapse, serial sampling and efficient longitudinal monitoring of disease progress and response to treatment. Various components of tumor cells released into the blood circulation can be analyzed in liquid biopsy sampling, some of which include circulating tumor cells (CTCs), circulating tumor DNA (ctDNA), cell-free RNA, tumor-educated platelets and exosomes. These components can be utilized for different purposes. As an example, ctDNA can be sequenced for genetic profiling of the tumors to enhance individualized treatment and longitudinal screening. CTC plasma count analysis or ctDNA detection after curative tumor resection surgery could facilitate early detection of minimal residual disease, aiding in the initiation of adjuvant therapy to prevent recurrence. Furthermore, CTC plasma count can be assessed to determine the stage and prognosis of breast cancer. In this review, we discuss the advantages and limitations of the various components of liquid biopsy used in breast cancer diagnosis and will expand on aspects that require further focus in future research.

Tamoxifen therapy benefit predictive signature combining with prognostic signature in surgical-only ER-positive breast cancer

Tamoxifen treatment is important assistant for estrogen-receptor-positive breast cancer (BRCA) after resection. This study aimed to identify signatures for predicting the prognosis of patients with BRCA after tamoxifen treatment. Data of gene-specific DNA methylation (DM), as well as the corresponding clinical data for the patients with BRCA, were obtained from The Cancer Genome Atlas and followed by systematic bioinformatics analyses. After mapping these DM CPG sites onto genes, we finally obtained 352 relapse-free survival (RFS) associated DM genes, with which 61,776 gene pairs were combined, including 1,614 gene pairs related to RFS. An 11 gene-pair signature was identified to cluster the 189 patients with BRCA into the surgical low-risk group (136 patients) and high-risk group (53 patients). Then, we further identified a tamoxifen-predictive signature that could classify surgical high-risk patients with significant differences on RFS. Combining surgical-only prognostic signature and tamoxifen-predictive signature, patients were clustered into surgical-only low-risk group, tamoxifen nonbenefit group, and tamoxifen benefit group. In conclusion, we identified that the gene pair PDHA2-APRT could serve as a potential prognostic biomarker for patients with BRCA after tamoxifen treatment.

Characterizing the replicability of cell types defined by single cell RNA-sequencing data using MetaNeighbor

Single-cell RNA-sequencing (scRNA-seq) technology provides a new avenue to discover and characterize cell types; however, the experiment-specific technical biases and analytic variability inherent to current pipelines may undermine its replicability. Meta-analysis is further hampered by the use of ad hoc naming conventions. Here we demonstrate our replication framework, MetaNeighbor, that quantifies the degree to which cell types replicate across datasets, and enables rapid identification of clusters with high similarity. We first measure the replicability of neuronal identity, comparing results across eight technically and biologically diverse datasets to define best practices for more complex assessments. We then apply this to novel interneuron subtypes, finding that 24/45 subtypes have evidence of replication, which enables the identification of robust candidate marker genes. Across tasks we find that large sets of variably expressed genes can identify replicable cell types with high accuracy, suggesting a general route forward for large-scale evaluation of scRNA-seq data.

Integrated analysis of differentially expressed genes and pathways in triple‑negative breast cancer

Triple‑negative breast cancer (TNBC) is a heterogeneous disease characterized by an aggressive phenotype and reduced survival. The aim of the present study was to investigate the molecular mechanisms involved in the carcinogenesis of TNBC and to identify novel target molecules for therapy. The differentially expressed genes (DEGs) in TNBC and normal adjacent tissue were assessed by analyzing the GSE41970 microarray data using Qlucore Omics Explorer, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes. Pathway enrichment analyses for DEGs were performed using the Database for Annotation, Visualization and Integrated Discovery online resource. A protein‑protein interaction (PPI) network was constructed using Search Tool for the Retrieval of Interacting Genes, and subnetworks were analyzed by ClusterONE. The PPI network and subnetworks were visualized using Cytoscape software. A total of 121 DEGs were obtained, of which 101 were upregulated and 20 were downregulated. The upregulated DEGs were significantly enriched in 14 pathways and 83 GO biological processes, while the downregulated DEGs were significantly enriched in 18 GO biological processes. The PPI network with 118 nodes and 1,264 edges was constructed and three subnetworks were extracted from the entire network. The significant hub DEGs with high degrees were identified, including TP53, glyceraldehyde‑3‑phosphate dehydrogenase, cyclin D1, HRAS and proliferating cell nuclear antigen, which were predominantly enriched in the cell cycle pathway and pathways in cancer. A number of critical genes and pathways were revealed to be associated with TNBC. The present study may provide an improved understanding of the pathogenesis of TNBC and contribute to the development of therapeutic targets for TNBC.

A functional signal profiling test for identifying a subset of HER2-negative breast cancers with abnormally amplified HER2 signaling activity

The results of clinical trials evaluating the efficacy of HER2 inhibitors in patients with breast cancer indicate that the correlation between HER2 receptor levels and patient outcomes is as low as 50%. The relatively weak correlation between HER2 status and response to HER2-targeting drugs suggests that measurement of HER2 signaling activity, rather than absolute HER2 levels, may more accurately diagnose HER2-driven breast cancer. A new diagnostic test, the CELx HER2 Signaling Profile (CELx HSP) test, is demonstrated to measure real-time HER2 signaling function in live primary cells. In the present study, epithelial cells extracted fresh from breast cancer patient tumors classified as HER2 negative (HER2-, n = 34 of which 33 were estrogen receptor positive) and healthy subjects (n = 16) were evaluated along with reference breast cancer cell lines (n = 19). Live cell response to specific HER2 agonists (NRG1b and EGF) and antagonist (pertuzumab) was measured. Of the HER2- breast tumor cell samples tested, 7 of 34 patients (20.5%; 95% CI = 10%-37%) had HER2 signaling activity that was characterized as abnormally high. Amongst the tumor samples there was no correlation between HER2 protein status (by cell cytometry) and HER2 signaling activity (hyperactive or normal) (Regression analysis P = 0.144, R2 = 0.068). One conclusion is that measurement of HER2 signaling activity can identify a subset of breast cancers with normal HER2 receptor levels with abnormally high levels of HER2 signaling. This result constitutes a new subtype of breast cancer that should be considered for treatment with HER2 pathway inhibitors.

Associations of mRNA:microRNA for the Shared Downstream Molecules of EGFR and Alternative Tyrosine Kinase Receptors in Non-small Cell Lung Cancer

Lung cancer is the top cancer killer worldwide with high mortality rate. Majority belong to non-small cell lung cancers (NSCLCs). The epidermal growth factor receptor (EGFR) has been broadly explored as a drug target for therapy. However, the drug responses are not durable due to the acquired resistance. MicroRNAs (miRNAs) are small non-coding and endogenous molecules that can inhibit mRNA translation initiation and degrade mRNAs. We wonder if some downstream molecules shared by EGFR and the other tyrosine kinase receptors (TKRs) further transduce the signals alternatively, and some miRNAs play the key roles in affecting the expression of these downstream molecules. In this study, we investigated the mRNA:miRNA associations for the direct EGFR downstream molecules in the EGFR signaling pathway shared with the other TKRs, including c-MET (hepatocyte growth factor receptor), Ron (a protein tyrosine kinase related to c-MET), PDGFR (platelet-derived growth factor receptor), and IGF-1R (insulin-like growth factor receptor-1). The multiple linear regression and support vector regression (SVR) models were used to discover the statistically significant and the best weighted miRNAs regulating the mRNAs of these downstream molecules. These two models revealed the similar mRNA:miRNA associations. It was found that the miRNAs significantly affecting the mRNA expressions in the multiple regression model were also those with the largest weights in the SVR model. To conclude, we effectively identified a list of meaningful mRNA:miRNA associations: phospholipase C, gamma 1 (PLCG1) with miR-34a, phosphoinositide-3-kinase, regulatory subunit 2 (PIK3R2) with miR-30a-5p, growth factor receptor-bound protein 2 (GRB2) with miR-27a, and Janus kinase 1 (JAK1) with miR-302b and miR-520e. These associations could make great contributions to explore new mechanism in NSCLCs. These candidate miRNAs may be regarded as the potential drug targets for treating NSCLCs with acquired drug resistance.

Co-expression Pattern Analysis of miR-17-92 Target Genes in Chronic Myelogenous Leukemia

MicroRNAs (miRNAs) are post-transcriptional regulators that regulate gene expression by binding to the 3' untranslated region of target mRNAs. Mature miRNAs transcribed from the miR-17-92 cluster have an oncogenic activity, which are overexpressed in chronic-phase chronic myelogenous leukemia (CML) patients compared with normal individuals. Besides, the tyrosine kinase activity of BCR-ABL oncoprotein from the Philadelphia chromosome in CML can affect this miRNA cluster. Genes with similar mRNA expression profiles are likely to be regulated by the same regulators. We hypothesize that target genes regulated by the same miRNA are co-expressed. In this study, we aim to explore the difference in the co-expression patterns of those genes potentially regulated by miR-17-92 cluster between the normal and the CML groups. We applied a statistical method for gene pair classification by identifying a disease-specific cutoff point that classified the co-expressed gene pairs into strong and weak co-expression classes. The method effectively identified the differences in the co-expression patterns from the overall structure. Functional annotation for co-expressed gene pairs showed that genes involved in the metabolism processes were more likely to be co-expressed in the normal group compared to the CML group. Our method can identify the co-expression pattern difference from the overall structure between two different distributions using the distribution-based statistical method. Functional annotation further provides the biological support. The co-expression pattern in the normal group is regarded as the inter-gene linkages, which represents the healthy pathological balance. Dysregulation of metabolism may be related to CML pathology. Our findings will provide useful information for investigating the novel CML mechanism and treatment.

Breast cancer subtype predictors revisited: from consensus to concordance?

BACKGROUND

At the molecular level breast cancer comprises a heterogeneous set of subtypes associated with clear differences in gene expression and clinical outcomes. Single sample predictors (SSPs) are built via a two-stage approach consisting of clustering and subtype predictor construction based on the cluster labels of individual cases. SSPs have been criticized because their subtype assignments for the same samples were only moderately concordant (Cohen's κ<0.6).

METHODS

We propose a semi-supervised approach where for five datasets, consensus sets were constructed consisting of those samples that were concordantly subtyped by a number of different predictors. Next, nine subtype predictors - three SSPs, three subtype classification models (SCMs) and three novel rule-based predictors based on the St. Gallen surrogate intrinsic subtype definitions (STGs) - were constructed on the five consensus sets and their associated consensus subtype labels. The predictors were validated on a compendium of over 4,000 uniformly preprocessed Affymetrix microarrays. Concordance between subtype predictors was assessed using Cohen's kappa statistic.

RESULTS

In this standardized setup, subtype predictors of the same type (either SCM, SSP, or STG) but with a different gene list and/or consensus training set were associated with almost perfect levels of agreement (median κ>0.8). Interestingly, for a given predictor type a change in consensus set led to higher concordance than a change to another gene list. The more challenging scenario where the predictor type, gene list and training set were all different resulted in predictors with only substantial levels of concordance (median κ=0.74) on independent validation data.

CONCLUSIONS

Our results demonstrate that for a given subtype predictor type stringent standardization of the preprocessing stage, combined with carefully devised consensus training sets, leads to predictors that show almost perfect levels of concordance. However, predictors of a different type are only substantially concordant, despite reaching almost perfect levels of concordance on training data.

Placental transcriptome in development and pathology: expression, function, and methods of analysis

The placenta is the essential organ of mammalian pregnancy and errors in its development and function are associated with a wide range of human pathologies of pregnancy. Genome sequencing has led to methods for investigation of the transcriptome (all expressed RNA species) using microarrays and next-generation sequencing, and implementation of these techniques has identified many novel species of RNA including: micro-RNA, long noncoding RNA, and circular RNA. These species can physically interact with both each other and regulatory proteins to modify gene expression and messenger RNA to protein translation. Transcriptome analysis is actively used to investigate placental development and dysfunction in pathologies ranging from preeclampsia and fetal growth restriction to preterm labor. Genome-wide gene expression analysis is also being applied to identify prognostic and diagnostic biomarkers of these disorders. In this comprehensive review we summarize transcriptome biology, methods of isolation and analysis, application to placental development and pathology, and use in diagnostic analysis in maternal blood. Key information for analysis methods is organized into quick reference tables where current analysis techniques and tools are cited and compared. We have created this review as a practical guide and starting reference for those interested in beginning an investigation into the transcriptome of the placenta.

Coexpression Pattern Analysis of NPM1-Associated Genes in Chronic Myelogenous Leukemia

Background. Nucleophosmin 1 (NPM1) plays an important role in ribosomal synthesis and malignancies, but NPM1 mutations occur rarely in the blast-crisis and chronic-phase chronic myelogenous leukemia (CML) patients. The NPM1-associated gene set (GCM_NPM1), in total 116 genes including NPM1, was chosen as the candidate gene set for the coexpression analysis. We wonder if NPM1-associated genes can affect the ribosomal synthesis and translation process in CML. Results. We presented a distribution-based approach for gene pair classification by identifying a disease-specific cutoff point that classified the coexpressed gene pairs into strong and weak coexpression structures. The differences in the coexpression patterns between the normal and the CML groups were reflected from the overall structure by performing two-sample Kolmogorov-Smirnov test. Our developed method effectively identified the coexpression pattern differences from the overall structure: P  value = 1.71 × 10⁻²² < 0.05 for the maximum deviation D = 0.109. Moreover, we found that genes involved in the ribosomal synthesis and translation process tended to be coexpressed in the CML group. Conclusion. Our developed method can identify the coexpression difference between two different groups. Dysregulation of ribosomal synthesis and translation process may be related to the CML disease. Our significant findings may provide useful information for the novel CML mechanism exploration and cancer treatment.

Molecular Classification of Breast Carcinoma: From Traditional, Old-Fashioned Way to A New Age, and A New Way

Breast carcinoma comprises a group of diseases with specific clinical, histopathologic and molecular properties. Traditional classification use morphology to divide tumors into separate categories with differing behavior and prognosis. However, there are limitations of traditional classification systems, and new molecular methods are expected to improve classification systems. Molecular subtypes of breast carcinomas have been characterized in the last 11 years, and have been studied extensively. Much of the information accumulated in recent years, and molecular taxonomy seems to be still developing and undergoing change. The main question is whether new molecular techniques such as gene expression profiling will be accepted as gold standard in determining breast cancer subtypes, and whether molecular classification is useful in specific subtypes of breast cancer as it is in ductal carcinoma (nonspecific type). In addition, critical review of the literature reveals major problems such as poor definition, lack of reproducibility and lack of quality control in current molecular techniques and classifications. Therefore, current molecular approaches are not yet used in routine clinical practice and treatment guidance since they are immature and can even lead to incorrect assessment.

Expression of estrogen-induced genes and estrogen receptor β in pancreatic neuroendocrine tumors: implications for targeted therapy

OBJECTIVE

The indolent nature and expression of progesterone receptor (PR), a well-known estrogen-induced gene, in a subset of pancreatic neuroendocrine tumors (PanNETs), raise the possibility of hormonal regulation in these tumors.

METHODS

Immunohistochemical expression of estrogen receptors (ERs) α and β as well as messenger RNA expression of estrogen-induced genes (PR, EIG121, IGF-1, IGF-1R, sFRP1, and sFRP4) by quantitative reverse transcription-polymerase chain reaction were examined in 131 World Health Organization grade G1 and G2 PanNETs and correlated their expression with clinicopathological features.

RESULTS

Thirty-nine PanNETs (30%) showed high positive ERβ staining, and 87 cases (66%) had low positive ERβ staining; only 5 cases (4%) had no nuclear staining. Pancreatic neuroendocrine tumors with small size (P = 0.02), low World Health Organization grade (P = 0.02), and low American Joint Committee on Cancer stage (P = 0.006) more frequently showed high positive ERβ staining. Among the estrogen-induced genes studied, PanNETs had significantly higher expression of PR, EIG121, IGF-1, sFRP1, and sFRP4 compared with normal pancreas, independent of age or sex. High positive ERβ staining was associated with an increased expression of PR (P < 0.001) and EIG121 (P = 0.02).

CONCLUSIONS

Our study showed that PanNETs with favorable prognostic features have higher ERβ expression, which is associated with up-regulated PR and EIG121 messenger RNA expression. Estrogen regulation in PanNETs could potentially help in risk stratification and provide a rational target for novel treatment strategies.

Novel approach for coexpression analysis of E2F1-3 and MYC target genes in chronic myelogenous leukemia

BACKGROUND

Chronic myelogenous leukemia (CML) is characterized by tremendous amount of immature myeloid cells in the blood circulation. E2F1-3 and MYC are important transcription factors that form positive feedback loops by reciprocal regulation in their own transcription processes. Since genes regulated by E2F1-3 or MYC are related to cell proliferation and apoptosis, we wonder if there exists difference in the coexpression patterns of genes regulated concurrently by E2F1-3 and MYC between the normal and the CML states.

RESULTS

We proposed a method to explore the difference in the coexpression patterns of those candidate target genes between the normal and the CML groups. A disease-specific cutoff point for coexpression levels that classified the coexpressed gene pairs into strong and weak coexpression classes was identified. Our developed method effectively identified the coexpression pattern differences from the overall structure. Moreover, we found that genes related to the cell adhesion and angiogenesis properties were more likely to be coexpressed in the normal group when compared to the CML group.

CONCLUSION

Our findings may be helpful in exploring the underlying mechanisms of CML and provide useful information in cancer treatment.

Breast carcinomas: variations in sonoelastographic appearance

Background

This study assessed factors influencing the sonoelastographic presentation of breast carcinoma.

Methods

A prospective collaborative study was conducted by the Santa Casa de São Paulo and CTC-Center, on 540 breast lesions in women referred for percutaneous breast biopsy. Eighty-four carcinomas showing lesions on ultrasonography were included. These lesions were classified into four sonoelastographic scores, where scores of 1, 2, and 3 were considered false-negative, and a score of 4 was considered true-positive. Scores were compared against histopathologic results, which were divided into two groups, ie, soft lesions (group 1) and hard lesions (group 2). False-negative and true-positive results were also assessed for variation according to patient age and mean lesion diameter.

Results

Of the 84 lesions studied, nine yielded false-negative results on sonoelastography and 75 yielded true-positive results. In terms of histopathologic classification, eight were assigned to group 1 and 76 to group 2. The chi-squared test showed a correlation between sonoelastographic scores and histopathologic lesion type. No statistically significant differences were observed according to patient age or largest lesion diameter.

Conclusion

Our results revealed an association between sonoelastographic presentation of breast lesions and histology. False-negative results on sonoelastography were influenced by histologic type of lesion and not by lesion size or patient age.

Multiple regression analysis of mRNA-miRNA associations in colorectal cancer pathway

Background. MicroRNA (miRNA) is a short and endogenous RNA molecule that regulates posttranscriptional gene expression. It is an important factor for tumorigenesis of colorectal cancer (CRC), and a potential biomarker for diagnosis, prognosis, and therapy of CRC. Our objective is to identify the related miRNAs and their associations with genes frequently involved in CRC microsatellite instability (MSI) and chromosomal instability (CIN) signaling pathways. Results. A regression model was adopted to identify the significantly associated miRNAs targeting a set of candidate genes frequently involved in colorectal cancer MSI and CIN pathways. Multiple linear regression analysis was used to construct the model and find the significant mRNA-miRNA associations. We identified three significantly associated mRNA-miRNA pairs: BCL2 was positively associated with miR-16 and SMAD4 was positively associated with miR-567 in the CRC tissue, while MSH6 was positively associated with miR-142-5p in the normal tissue. As for the whole model, BCL2 and SMAD4 models were not significant, and MSH6 model was significant. The significant associations were different in the normal and the CRC tissues. Conclusion. Our results have laid down a solid foundation in exploration of novel CRC mechanisms, and identification of miRNA roles as oncomirs or tumor suppressor mirs in CRC.

A personalized committee classification approach to improving prediction of breast cancer metastasis

MOTIVATION

Metastasis prediction is a well-known problem in breast cancer research. As breast cancer is a complex and heterogeneous disease with many molecular subtypes, predictive models trained for one cohort often perform poorly on other cohorts, and a combined model may be suboptimal for individual patients. Furthermore, attempting to develop subtype-specific models is hindered by the ambiguity and stereotypical definitions of subtypes.

RESULTS

Here, we propose a personalized approach by relaxing the definition of breast cancer subtypes. We assume that each patient belongs to a distinct subtype, defined implicitly by a set of patients with similar molecular characteristics, and construct a different predictive model for each patient, using as training data, only the patients defining the subtype. To increase robustness, we also develop a committee-based prediction method by pooling together multiple personalized models. Using both intra- and inter-dataset validations, we show that our approach can significantly improve the prediction accuracy of breast cancer metastasis compared with several popular approaches, especially on those hard-to-learn cases. Furthermore, we find that breast cancer patients belonging to different canonical subtypes tend to have different predictive models and gene signatures, suggesting that metastasis in different canonical subtypes are likely governed by different molecular mechanisms.

AVAILABILITY AND IMPLEMENTATION

Source code implemented in MATLAB and Java available at www.cs.utsa.edu/∼jruan/PCC/.

Biclustering reveals breast cancer tumour subgroups with common clinical features and improves prediction of disease recurrence

BACKGROUND

Many studies have revealed correlations between breast tumour phenotypes, variations in gene expression, and patient survival outcomes. The molecular heterogeneity between breast tumours revealed by these studies has allowed prediction of prognosis and has underpinned stratified therapy, where groups of patients with particular tumour types receive specific treatments. The molecular tests used to predict prognosis and stratify treatment usually utilise fixed sets of genomic biomarkers, with the same biomarker sets being used to test all patients. In this paper we suggest that instead of fixed sets of genomic biomarkers, it may be more effective to use a stratified biomarker approach, where optimal biomarker sets are automatically chosen for particular patient groups, analogous to the choice of optimal treatments for groups of similar patients in stratified therapy. We illustrate the effectiveness of a biclustering approach to select optimal gene sets for determining the prognosis of specific strata of patients, based on potentially overlapping, non-discrete molecular characteristics of tumours.

RESULTS

Biclustering identified tightly co-expressed gene sets in the tumours of restricted subgroups of breast cancer patients. The co-expressed genes in these biclusters were significantly enriched for particular biological annotations and gene regulatory modules associated with breast cancer biology. Tumours identified within the same bicluster were more likely to present with similar clinical features. Bicluster membership combined with clinical information could predict patient prognosis in conditional inference tree and ridge regression class prediction models.

CONCLUSIONS

The increasing clinical use of genomic profiling demands identification of more effective methods to segregate patients into prognostic and treatment groups. We have shown that biclustering can be used to select optimal gene sets for determining the prognosis of specific strata of patients.

Circulating Tumor Cells Detected by RT-PCR for CK-20 before Surgery Indicate Worse Prognostic Impact in Triple-Negative and HER2 Subtype Breast Cancer

Purpose

Circulating tumor cells (CTC) clearly correlate with unfavorable outcomes for patients with metastatic breast cancer, but the long-term prognostic implications of CTC for molecular subtypes of operable breast cancer are not yet known. We explored the relationships between previously established prognostic factors and CTC in operable breast cancer, and the significance of CTC by breast cancer molecular subtype.

Methods

We retrospectively evaluated 166 patients with operable breast cancer (stage I-IIIA) diagnosed from April 1997 to May 2003. CTC were detected using cytokeratin-20 (CK-20) mRNA expression in peripheral blood samples that were collected just prior to surgery under general anesthesia. Clinicopathological characteristics of the cancer were analyzed according to CTC status. Metastasis-free survival (MFS) and overall survival (OS) were analyzed according to CTC status and breast cancer molecular subtype.

Results

CK-20 mRNA-positive CTC was detected in 37 of 166 patients (22.3%) and was not correlated with any previous clinical factors in univariate analysis (p>0.05). After a median follow-up of 100 months, the patients with CK-20 mRNA-positive CTC had less favorable outcomes in terms of MFS and OS than those without detectable CTC (log-rank p<0.05). Among molecular subtypes of operable breast cancer, the patients with CK-20 mRNA-positive CTC had shorter MFS and OS in triple negative and human epidermal growth factor 2 (HER2) breast cancer subtype (log-rank, p<0.05).

Conclusion

CK-20 mRNA-positive CTC may lend insight into tumor progression as a prognostic indicator especially in the triple negative and HER2 subtypes of operable breast cancer.

Angiogenic mRNA and microRNA gene expression signature predicts a novel subtype of serous ovarian cancer

Ovarian cancer is the fifth leading cause of cancer death for women in the U.S. and the seventh most fatal worldwide. Although ovarian cancer is notable for its initial sensitivity to platinum-based therapies, the vast majority of patients eventually develop recurrent cancer and succumb to increasingly platinum-resistant disease. Modern, targeted cancer drugs intervene in cell signaling, and identifying key disease mechanisms and pathways would greatly advance our treatment abilities. In order to shed light on the molecular diversity of ovarian cancer, we performed comprehensive transcriptional profiling on 129 advanced stage, high grade serous ovarian cancers. We implemented a, re-sampling based version of the ISIS class discovery algorithm (rISIS: robust ISIS) and applied it to the entire set of ovarian cancer transcriptional profiles. rISIS identified a previously undescribed patient stratification, further supported by micro-RNA expression profiles, and gene set enrichment analysis found strong biological support for the stratification by extracellular matrix, cell adhesion, and angiogenesis genes. The corresponding "angiogenesis signature" was validated in ten published independent ovarian cancer gene expression datasets and is significantly associated with overall survival. The subtypes we have defined are of potential translational interest as they may be relevant for identifying patients who may benefit from the addition of anti-angiogenic therapies that are now being tested in clinical trials.

A three-gene model to robustly identify breast cancer molecular subtypes

BACKGROUND

Single sample predictors (SSPs) and Subtype classification models (SCMs) are gene expression-based classifiers used to identify the four primary molecular subtypes of breast cancer (basal-like, HER2-enriched, luminal A, and luminal B). SSPs use hierarchical clustering, followed by nearest centroid classification, based on large sets of tumor-intrinsic genes. SCMs use a mixture of Gaussian distributions based on sets of genes with expression specifically correlated with three key breast cancer genes (estrogen receptor [ER], HER2, and aurora kinase A [AURKA]). The aim of this study was to compare the robustness, classification concordance, and prognostic value of these classifiers with those of a simplified three-gene SCM in a large compendium of microarray datasets.

METHODS

Thirty-six publicly available breast cancer datasets (n = 5715) were subjected to molecular subtyping using five published classifiers (three SSPs and two SCMs) and SCMGENE, the new three-gene (ER, HER2, and AURKA) SCM. We used the prediction strength statistic to estimate robustness of the classification models, defined as the capacity of a classifier to assign the same tumors to the same subtypes independently of the dataset used to fit it. We used Cohen κ and Cramer V coefficients to assess concordance between the subtype classifiers and association with clinical variables, respectively. We used Kaplan-Meier survival curves and cross-validated partial likelihood to compare prognostic value of the resulting classifications. All statistical tests were two-sided.

RESULTS

SCMs were statistically significantly more robust than SSPs, with SCMGENE being the most robust because of its simplicity. SCMGENE was statistically significantly concordant with published SCMs (κ = 0.65-0.70) and SSPs (κ = 0.34-0.59), statistically significantly associated with ER (V = 0.64), HER2 (V = 0.52) status, and histological grade (V = 0.55), and yielded similar strong prognostic value.

CONCLUSION

Our results suggest that adequate classification of the major and clinically relevant molecular subtypes of breast cancer can be robustly achieved with quantitative measurements of three key genes.

Basal breast cancer: a complex and deadly molecular subtype

During the last decade, gene expression profiling of breast cancer has revealed the existence of five molecular subtypes and allowed the establishment of a new classification. The basal subtype, which represents 15-25% of cases, is characterized by an expression profile similar to that of myoepithelial normal mammary cells. Basal tumors are frequently assimilated to triple-negative (TN) breast cancers. They display epidemiological and clinico-pathological features distinct from other subtypes. Their pattern of relapse is characterized by frequent and early relapses and visceral locations. Despite a relative sensitivity to chemotherapy, the prognosis is poor. Recent characterization of their molecular features, such as the dysfunction of the BRCA1 pathway or the frequent expression of EGFR, provides opportunities for optimizing the systemic treatment. Several clinical trials dedicated to basal or TN tumors are testing cytotoxic agents and/or molecularly targeted therapies. This review summarizes the current state of knowledge of this aggressive and hard-to-treat subtype of breast cancer.

Predictive and prognostic molecular markers for cancer medicine

Over the last 10 years there has been an explosion of information about the molecular biology of cancer. A challenge in oncology is to translate this information into advances in patient care. While there are well-formed routes for translating new molecular information into drug therapy, the routes for translating new information into sensitive and specific diagnostic, prognostic and predictive tests are still being developed. Similarly, the science of using tumor molecular profiles to select clinical trial participants or to optimize therapy for individual patients is still in its infancy. This review will summarize the current technologies for predicting treatment response and prognosis in cancer medicine, and outline what the future may hold. It will also highlight the potential importance of methods that can integrate molecular, histopathological and clinical information into a synergistic understanding of tumor progression. While these possibilities are without doubt exciting, significant challenges remain if we are to implement them with a strong evidence base in a widely available and cost-effective manner.

Low penetrance breast cancer susceptibility loci are associated with specific breast tumor subtypes: findings from the Breast Cancer Association Consortium

Breast cancers demonstrate substantial biological, clinical and etiological heterogeneity. We investigated breast cancer risk associations of eight susceptibility loci identified in GWAS and two putative susceptibility loci in candidate genes in relation to specific breast tumor subtypes. Subtypes were defined by five markers (ER, PR, HER2, CK5/6, EGFR) and other pathological and clinical features. Analyses included up to 30 040 invasive breast cancer cases and 53 692 controls from 31 studies within the Breast Cancer Association Consortium. We confirmed previous reports of stronger associations with ER+ than ER- tumors for six of the eight loci identified in GWAS: rs2981582 (10q26) (P-heterogeneity = 6.1 × 10(-18)), rs3803662 (16q12) (P = 3.7 × 10(-5)), rs13281615 (8q24) (P = 0.002), rs13387042 (2q35) (P = 0.006), rs4973768 (3p24) (P = 0.003) and rs6504950 (17q23) (P = 0.002). The two candidate loci, CASP8 (rs1045485, rs17468277) and TGFB1 (rs1982073), were most strongly related with the risk of PR negative tumors (P = 5.1 × 10(-6) and P = 4.1 × 10(-4), respectively), as previously suggested. Four of the eight loci identified in GWAS were associated with triple negative tumors (P ≤ 0.016): rs3803662 (16q12), rs889312 (5q11), rs3817198 (11p15) and rs13387042 (2q35); however, only two of them (16q12 and 2q35) were associated with tumors with the core basal phenotype (P ≤ 0.002). These analyses are consistent with different biological origins of breast cancers, and indicate that tumor stratification might help in the identification and characterization of novel risk factors for breast cancer subtypes. This may eventually result in further improvements in prevention, early detection and treatment.

Basal phenotype breast cancer: implications for treatment and prognosis

Breast cancer is the most common malignancy in females. The origins and biology of breast carcinomas remain unclear. Cellular and molecular heterogeneity results in different distinct groups of tumors with different clinical behavior and prognosis. Gene expression profiling has delineated five molecular subtypes based on similarities in gene expression: luminal A, luminal B, HER2 overexpressing, normal-like and basal-like. Basal-like breast cancer (BLBC) lacks estrogen receptor, progesterone receptor and HER2 expression, and comprises myoepithelial cells. Specific features include high proliferative rate, rapid growth, early recurrence and decreased overall survival. BLBC is associated with ductal carcinoma in situ, BRCA1 mutation, brain and lung metastasis, and negative axillary lymph nodes. Currently, chemotherapy is the only therapeutic choice, but demonstrates poor outcomes. There is an overlap in definition between triple-negative breast cancer and BLBC due to the triple-negative profile of BLBC. Despite the molecular and clinical similarities, the two subtypes respond differently to neoadjuvant therapy. Although particular morphologic, genetic and clinical features of BLBC have been identified, a variety of definitions among studies accounts for the contradictory results reported. In this article the molecular morphological and histopathological profile, the clinical behavior and the therapeutic options of BLBC are presented, with emphasis on the discordant findings among studies.

A refined molecular taxonomy of breast cancer

The current histoclinical breast cancer classification is simple but imprecise. Several molecular classifications of breast cancers based on expression profiling have been proposed as alternatives. However, their reliability and clinical utility have been repeatedly questioned, notably because most of them were derived from relatively small initial patient populations. We analyzed the transcriptomes of 537 breast tumors using three unsupervised classification methods. A core subset of 355 tumors was assigned to six clusters by all three methods. These six subgroups overlapped with previously defined molecular classes of breast cancer, but also showed important differences, notably the absence of an ERBB2 subgroup and the division of the large luminal ER+ group into four subgroups, two of them being highly proliferative. Of the six subgroups, four were ER+/PR+/AR+, one was ER−/PR−/AR+ and one was triple negative (AR−/ER−/PR−). ERBB2-amplified tumors were split between the ER−/PR−/AR+ subgroup and the highly proliferative ER+ LumC subgroup. Importantly, each of these six molecular subgroups showed specific copy-number alterations. Gene expression changes were correlated to specific signaling pathways. Each of these six subgroups showed very significant differences in tumor grade, metastatic sites, relapse-free survival or response to chemotherapy. All these findings were validated on large external datasets including more than 3000 tumors. Our data thus indicate that these six molecular subgroups represent well-defined clinico-biological entities of breast cancer. Their identification should facilitate the detection of novel prognostic factors or therapeutical targets in breast cancer.

An evaluation protocol for subtype-specific breast cancer event prediction

In recent years increasing evidence appeared that breast cancer may not constitute a single disease at the molecular level, but comprises a heterogeneous set of subtypes. This suggests that instead of building a single monolithic predictor, better predictors might be constructed that solely target samples of a designated subtype, which are believed to represent more homogeneous sets of samples. An unavoidable drawback of developing subtype-specific predictors, however, is that a stratification by subtype drastically reduces the number of samples available for their construction. As numerous studies have indicated sample size to be an important factor in predictor construction, it is therefore questionable whether the potential benefit of subtyping can outweigh the drawback of a severe loss in sample size. Factors like unequal class distributions and differences in the number of samples per subtype, further complicate comparisons. We present a novel experimental protocol that facilitates a comprehensive comparison between subtype-specific predictors and predictors that do not take subtype information into account. Emphasis lies on careful control of sample size as well as class and subtype distributions. The methodology is applied to a large breast cancer compendium involving over 1500 arrays, using a state-of-the-art subtyping scheme. We show that the resulting subtype-specific predictors outperform those that do not take subtype information into account, especially when taking sample size considerations into account.

A comparison of PAM50 intrinsic subtyping with immunohistochemistry and clinical prognostic factors in tamoxifen-treated estrogen receptor-positive breast cancer

PURPOSE

To compare clinical, immunohistochemical (IHC), and gene expression models of prognosis applicable to formalin-fixed, paraffin-embedded blocks in a large series of estrogen receptor (ER)-positive breast cancers from patients uniformly treated with adjuvant tamoxifen.

EXPERIMENTAL DESIGN

Quantitative real-time reverse transcription-PCR (qRT-PCR) assays for 50 genes identifying intrinsic breast cancer subtypes were completed on 786 specimens linked to clinical (median follow-up, 11.7 years) and IHC [ER, progesterone receptor (PR), HER2, and Ki67] data. Performance of predefined intrinsic subtype and risk-of-relapse scores was assessed using multivariable Cox models and Kaplan-Meier analysis. Harrell's C-index was used to compare fixed models trained in independent data sets, including proliferation signatures.

RESULTS

Despite clinical ER positivity, 10% of cases were assigned to nonluminal subtypes. qRT-PCR signatures for proliferation genes gave more prognostic information than clinical assays for hormone receptors or Ki67. In Cox models incorporating standard prognostic variables, hazard ratios for breast cancer disease-specific survival over the first 5 years of follow-up, relative to the most common luminal A subtype, are 1.99 [95% confidence interval (CI), 1.09-3.64] for luminal B, 3.65 (95% CI, 1.64-8.16) for HER2-enriched subtype, and 17.71 (95% CI, 1.71-183.33) for the basal-like subtype. For node-negative disease, PAM50 qRT-PCR-based risk assignment weighted for tumor size and proliferation identifies a group with >95% 10-year survival without chemotherapy. In node-positive disease, PAM50-based prognostic models were also superior.

CONCLUSION

The PAM50 gene expression test for intrinsic biological subtype can be applied to large series of formalin-fixed, paraffin-embedded breast cancers, and gives more prognostic information than clinical factors and IHC using standard cut points.

LSOSS: Detection of Cancer Outlier Differential Gene Expression

Detection of differential gene expression using microarray technology has received considerable interest in cancer research studies. Recently, many researchers discovered that oncogenes may be activated in some but not all samples in a given disease group. The existing statistical tools for detecting differentially expressed genes in a subset of the disease group mainly include cancer outlier profile analysis (COPA), outlier sum (OS), outlier robust t-statistic (ORT) and maximum ordered subset t-statistics (MOST). In this study, another approach named Least Sum of Ordered Subset Square t-statistic (LSOSS) is proposed. The results of our simulation studies indicated that LSOSS often has more power than previous statistical methods. When applied to real human breast and prostate cancer data sets, LSOSS was competitive in terms of the biological relevance of top ranked genes. Furthermore, a modified hierarchical clustering method was developed to classify the heterogeneous gene activation patterns of human breast cancer samples based on the significant genes detected by LSOSS. Three classes of gene activation patterns, which correspond to estrogen receptor (ER)+, ER− and a mixture of ER+ and ER−, were detected and each class was assigned a different gene signature.

A differentiation-based phylogeny of cancer subtypes

Histopathological classification of human tumors relies in part on the degree of differentiation of the tumor sample. To date, there is no objective systematic method to categorize tumor subtypes by maturation. In this paper, we introduce a novel computational algorithm to rank tumor subtypes according to the dissimilarity of their gene expression from that of stem cells and fully differentiated tissue, and thereby construct a phylogenetic tree of cancer. We validate our methodology with expression data of leukemia, breast cancer and liposarcoma subtypes and then apply it to a broader group of sarcomas. This ranking of tumor subtypes resulting from the application of our methodology allows the identification of genes correlated with differentiation and may help to identify novel therapeutic targets. Our algorithm represents the first phylogeny-based tool to analyze the differentiation status of human tumors.

Gene expression profiling of malignancies is often held to demonstrate genes that are “up-regulated” or “down-regulated”, but the appropriate frame of reference against which observations should be compared has not been determined. Fully differentiated somatic cells arise from stem cells, with changes in gene expression that can be experimentally determined. If cancers arise as the result of an abruption of the differentiation process, then poorly differentiated cancers would have a gene expression more similar to stem cells than to normal differentiated tissue, and well differentiated cancers would have a gene expression more similar to fully differentiated cells than to stem cells. In this paper, we describe a novel computational algorithm that allows orientation of cancer gene expression between the poles of the gene expression of stem cells and of fully differentiated tissue. Our methodology allows the construction of a multi-branched phylogeny of human malignancies and can be used to identify genes related to differentiation as well as novel therapeutic targets.

Gene expression in histologically normal epithelium from breast cancer patients and from cancer-free prophylactic mastectomy patients shares a similar profile

INTRODUCTION

We hypothesised that gene expression in histologically normal (HN) epithelium (NlEpi) would differ between breast cancer patients and usual-risk controls undergoing reduction mammoplasty (RM), and that gene expression in NlEpi from cancer-free prophylactic mastectomy (PM) samples from high-risk women would resemble HN gene expression.

METHODS

We analysed gene expression in 73 NlEpi samples microdissected from frozen tissue. In 42 samples, we used microarrays to compare gene expression between 18 RM patients and 18 age-matched HN (9 oestrogen receptor (ER)+, 9 ER-) and 6 PM patients. Data were analysed using a Bayesian approach (BADGE), and validated with quantitative real-time PCR (qPCR) in 31 independent NlEpi samples from 8 RM, 17 HN, and 6 PM patients.

RESULTS

A total of 98 probe sets (86 genes) were differentially expressed between RM and HN samples. Performing hierarchical analysis with these 98 probe sets, PM and HN samples clustered together, away from RM samples. qPCR validation of independent samples was high (84%) and uniform in RM compared with HN patients, and lower (58%), but more heterogeneous, in RM compared with PM patients. The 86 genes were implicated in many processes including transcription and the MAPK pathway.

CONCLUSION

Gene expression differs between the NlEpi of breast cancer cases and controls. The profile of cancer cases can be discerned in high-risk NlEpi from cancer-free breasts. This suggests that the profile is not an effect of the tumour, but may mark increased risk and reveal the earliest genomic changes of breast cancer.

Regularized Multivariate Regression for Identifying Master Predictors with Application to Integrative Genomics Study of Breast Cancer

In this paper, we propose a new method remMap - REgularized Multivariate regression for identifying MAster Predictors - for fitting multivariate response regression models under the high-dimension-low-sample-size setting. remMap is motivated by investigating the regulatory relationships among different biological molecules based on multiple types of high dimensional genomic data. Particularly, we are interested in studying the influence of DNA copy number alterations on RNA transcript levels. For this purpose, we model the dependence of the RNA expression levels on DNA copy numbers through multivariate linear regressions and utilize proper regularization to deal with the high dimensionality as well as to incorporate desired network structures. Criteria for selecting the tuning parameters are also discussed. The performance of the proposed method is illustrated through extensive simulation studies. Finally, remMap is applied to a breast cancer study, in which genome wide RNA transcript levels and DNA copy numbers were measured for 172 tumor samples. We identify a trans-hub region in cytoband 17q12-q21, whose amplification influences the RNA expression levels of more than 30 unlinked genes. These findings may lead to a better understanding of breast cancer pathology.

A fuzzy gene expression-based computational approach improves breast cancer prognostication

A fuzzy computational approach that takes into account several molecular subtypes in order to provide more accurate breast cancer prognosis

Early gene expression studies classified breast tumors into at least three clinically relevant subtypes. Although most current gene signatures are prognostic for estrogen receptor (ER) positive/human epidermal growth factor receptor 2 (HER2) negative breast cancers, few are informative for ER negative/HER2 negative and HER2 positive subtypes. Here we present Gene Expression Prognostic Index Using Subtypes (GENIUS), a fuzzy approach for prognostication that takes into account the molecular heterogeneity of breast cancer. In systematic evaluations, GENIUS significantly outperformed current gene signatures and clinical indices in the global population of patients.

A particle swarm based hybrid system for imbalanced medical data sampling

BACKGROUND

Medical and biological data are commonly with small sample size, missing values, and most importantly, imbalanced class distribution. In this study we propose a particle swarm based hybrid system for remedying the class imbalance problem in medical and biological data mining. This hybrid system combines the particle swarm optimization (PSO) algorithm with multiple classifiers and evaluation metrics for evaluation fusion. Samples from the majority class are ranked using multiple objectives according to their merit in compensating the class imbalance, and then combined with the minority class to form a balanced dataset.

RESULTS

One important finding of this study is that different classifiers and metrics often provide different evaluation results. Nevertheless, the proposed hybrid system demonstrates consistent improvements over several alternative methods with three different metrics. The sampling results also demonstrate good generalization on different types of classification algorithms, indicating the advantage of information fusion applied in the hybrid system.

CONCLUSION

The experimental results demonstrate that unlike many currently available methods which often perform unevenly with different datasets the proposed hybrid system has a better generalization property which alleviates the method-data dependency problem. From the biological perspective, the system provides indication for further investigation of the highly ranked samples, which may result in the discovery of new conditions or disease subtypes.

Molecular markers of breast axillary lymph node metastasis

In breast cancer, axillary lymph node status is one of the most important prognostic variables and a crucial component to the staging system. Several clinico-histopathological parameters are considered to be strong predictors of metastasis; however, they fail to accurately classify breast tumors according to their clinical behavior and to predict which patients will have disease recurrence. Methods based on genome-wide microarray analyses have been used to identify molecular markers with respect to the development of axillary lymph node metastasis. Most of these markers can be detected in the primary tumors, which can potentially lead to the ability to identify patients at the time of diagnosis who are at high risk for lymph node metastasis, allowing for early intervention and more suitable adjuvant treatments.

Evaluation of cDNA microarray data by multiple clones mapping to the same transcript

Although novel technologies are rapidly emerging, the cDNA microarray data accumulated is still and will be an important source for bioinformatics and biological studies. Thus, the reliability and applicability of the cDNA microarray data warrants further evaluation. In cDNA microarrays, multiple clones are measured for a transcript, which can be exploited to evaluate the consistency of microarray data. We show that even for pairs of RCs, the average Pearson correlation coefficient of their measurements is not high. However, this low consistency could largely be explained by random noise signals for a fraction of unexpressed genes and/or low signal-to-noise ratios for low abundance transcripts. Encouragingly, a large fraction of inconsistent data will be filtered out in the procedure of selecting differentially expressed genes (DEGs). Therefore, although cDNA microarray data are of low consistency, applications based on DEGs selections could still reach correct biological results, especially at the functional modules level.