A consensus prognostic gene expression classifier for ER positive breast cancer

A comprehensive analysis of prognostic signatures reveals the high predictive capacity of the proliferation, immune response and RNA splicing modules in breast cancer

Introduction

Several gene expression signatures have been proposed and demonstrated to be predictive of outcome in breast cancer. In the present article we address the following issues: Do these signatures perform similarly? Are there (common) molecular processes reported by these signatures? Can better prognostic predictors be constructed based on these identified molecular processes?

Methods

We performed a comprehensive analysis of the performance of nine gene expression signatures on seven different breast cancer datasets. To better characterize the functional processes associated with these signatures, we enlarged each signature by including all probes with a significant correlation to at least one of the genes in the original signature. The enrichment of functional groups was assessed using four ontology databases.

Results

The classification performance of the nine gene expression signatures is very similar in terms of assigning a sample to either a poor outcome group or a good outcome group. Nevertheless the concordance in classification at the sample level is low, with only 50% of the breast cancer samples classified in the same outcome group by all classifiers. The predictive accuracy decreases with the number of poor outcome assignments given to a sample. The best classification performance was obtained for the group of patients with only good outcome assignments. Enrichment analysis of the enlarged signatures revealed 11 functional modules with prognostic ability. The combination of the RNA-splicing and immune modules resulted in a classifier with high prognostic performance on an independent validation set.

Conclusions

The study revealed that the nine signatures perform similarly but exhibit a large degree of discordance in prognostic group assignment. Functional analyses indicate that proliferation is a common cellular process, but that other functional categories are also enriched and show independent prognostic ability. We provide new evidence of the potentially promising prognostic impact of immunity and RNA-splicing processes in breast cancer.

Prognostic breast cancer signature identified from 3D culture model accurately predicts clinical outcome across independent datasets

Background

One of the major tenets in breast cancer research is that early detection is vital for patient survival by increasing treatment options. To that end, we have previously used a novel unsupervised approach to identify a set of genes whose expression predicts prognosis of breast cancer patients. The predictive genes were selected in a well-defined three dimensional (3D) cell culture model of non-malignant human mammary epithelial cell morphogenesis as down-regulated during breast epithelial cell acinar formation and cell cycle arrest. Here we examine the ability of this gene signature (3D-signature) to predict prognosis in three independent breast cancer microarray datasets having 295, 286, and 118 samples, respectively.

Methods and Findings

Our results show that the 3D-signature accurately predicts prognosis in three unrelated patient datasets. At 10 years, the probability of positive outcome was 52, 51, and 47 percent in the group with a poor-prognosis signature and 91, 75, and 71 percent in the group with a good-prognosis signature for the three datasets, respectively (Kaplan-Meier survival analysis, p<0.05). Hazard ratios for poor outcome were 5.5 (95% CI 3.0 to 12.2, p<0.0001), 2.4 (95% CI 1.6 to 3.6, p<0.0001) and 1.9 (95% CI 1.1 to 3.2, p = 0.016) and remained significant for the two larger datasets when corrected for estrogen receptor (ER) status. Hence the 3D-signature accurately predicts breast cancer outcome in both ER-positive and ER-negative tumors, though individual genes differed in their prognostic ability in the two subtypes. Genes that were prognostic in ER+ patients are AURKA, CEP55, RRM2, EPHA2, FGFBP1, and VRK1, while genes prognostic in ER− patients include ACTB, FOXM1 and SERPINE2 (Kaplan-Meier p<0.05). Multivariable Cox regression analysis in the largest dataset showed that the 3D-signature was a strong independent factor in predicting breast cancer outcome.

Conclusions

The 3D-signature accurately predicts breast cancer outcome across multiple datasets and holds prognostic value for both ER-positive and ER-negative breast cancer. The signature was selected using a novel biological approach and hence holds promise to represent the key biological processes of breast cancer.

Pooling breast cancer datasets has a synergetic effect on classification performance and improves signature stability

Background

Michiels et al. (Lancet 2005; 365: 488–92) employed a resampling strategy to show that the genes identified as predictors of prognosis from resamplings of a single gene expression dataset are highly variable. The genes most frequently identified in the separate resamplings were put forward as a 'gold standard'. On a higher level, breast cancer datasets collected by different institutions can be considered as resamplings from the underlying breast cancer population. The limited overlap between published prognostic signatures confirms the trend of signature instability identified by the resampling strategy. Six breast cancer datasets, totaling 947 samples, all measured on the Affymetrix platform, are currently available. This provides a unique opportunity to employ a substantial dataset to investigate the effects of pooling datasets on classifier accuracy, signature stability and enrichment of functional categories.

Results

We show that the resampling strategy produces a suboptimal ranking of genes, which can not be considered to be a 'gold standard'. When pooling breast cancer datasets, we observed a synergetic effect on the classification performance in 73% of the cases. We also observe a significant positive correlation between the number of datasets that is pooled, the validation performance, the number of genes selected, and the enrichment of specific functional categories. In addition, we have evaluated the support for five explanations that have been postulated for the limited overlap of signatures.

Conclusion

The limited overlap of current signature genes can be attributed to small sample size. Pooling datasets results in more accurate classification and a convergence of signature genes. We therefore advocate the analysis of new data within the context of a compendium, rather than analysis in isolation.

The humoral immune system has a key prognostic impact in node-negative breast cancer

Estrogen receptor (ER) expression and proliferative activity are established prognostic factors in breast cancer. In a search for additional prognostic motifs, we analyzed the gene expression patterns of 200 tumors of patients who were not treated by systemic therapy after surgery using a discovery approach. After performing hierarchical cluster analysis, we identified coregulated genes related to the biological process of proliferation, steroid hormone receptor expression, as well as B-cell and T-cell infiltration. We calculated metagenes as a surrogate for all genes contained within a particular cluster and visualized the relative expression in relation to time to metastasis with principal component analysis. Distinct patterns led to the hypothesis of a prognostic role of the immune system in tumors with high expression of proliferation-associated genes. In multivariate Cox regression analysis, the proliferation metagene showed a significant association with metastasis-free survival of the whole discovery cohort [hazard ratio (HR), 2.20; 95% confidence interval (95% CI), 1.40-3.46]. The B-cell metagene showed additional independent prognostic information in carcinomas with high proliferative activity (HR, 0.66; 95% CI, 0.46-0.97). A prognostic influence of the B-cell metagene was independently confirmed by multivariate analysis in a first validation cohort enriched for high-grade tumors (n = 286; HR, 0.78; 95% CI, 0.62-0.98) and a second validation cohort enriched for younger patients (n = 302; HR, 0.83; 95% CI, 0.7-0.97). Thus, we could show in three cohorts of untreated, node-negative breast cancer patients that the humoral immune system plays a pivotal role in metastasis-free survival of carcinomas of the breast.

Meta-analysis of gene expression profiles in breast cancer: toward a unified understanding of breast cancer subtyping and prognosis signatures

Introduction

Breast cancer subtyping and prognosis have been studied extensively by gene expression profiling, resulting in disparate signatures with little overlap in their constituent genes. Although a previous study demonstrated a prognostic concordance among gene expression signatures, it was limited to only one dataset and did not fully elucidate how the different genes were related to one another nor did it examine the contribution of well-known biological processes of breast cancer tumorigenesis to their prognostic performance.

Method

To address the above issues and to further validate these initial findings, we performed the largest meta-analysis of publicly available breast cancer gene expression and clinical data, which are comprised of 2,833 breast tumors. Gene coexpression modules of three key biological processes in breast cancer (namely, proliferation, estrogen receptor [ER], and HER2 signaling) were used to dissect the role of constituent genes of nine prognostic signatures.

Results

Using a meta-analytical approach, we consolidated the signatures associated with ER signaling, ERBB2 amplification, and proliferation. Previously published expression-based nomenclature of breast cancer 'intrinsic' subtypes can be mapped to the three modules, namely, the ER^-/HER2^- (basal-like), the HER2⁺ (HER2-like), and the low- and high-proliferation ER⁺/HER2^- subtypes (luminal A and B). We showed that all nine prognostic signatures exhibited a similar prognostic performance in the entire dataset. Their prognostic abilities are due mostly to the detection of proliferation activity. Although ER^- status (basal-like) and ERBB2⁺ expression status correspond to bad outcome, they seem to act through elevated expression of proliferation genes and thus contain only indirect information about prognosis. Clinical variables measuring the extent of tumor progression, such as tumor size and nodal status, still add independent prognostic information to proliferation genes.

Conclusion

This meta-analysis unifies various results of previous gene expression studies in breast cancer. It reveals connections between traditional prognostic factors, expression-based subtyping, and prognostic signatures, highlighting the important role of proliferation in breast cancer prognosis.

Meta-analysis confirms BCL2 is an independent prognostic marker in breast cancer

Background

A number of protein markers have been investigated as prognostic adjuncts in breast cancer but their translation into clinical practice has been impeded by a lack of appropriate validation. Recently, we showed that BCL2 protein expression had prognostic power independent of current used standards. Here, we present the results of a meta-analysis of the association between BCL2 expression and both disease free survival (DFS) and overall survival (OS) in female breast cancer.

Methods

Reports published in 1994–2006 were selected for the meta-analysis using a search of PubMed. Studies that investigated the role of BCL2 expression by immunohistochemistry with a sample size greater than 100 were included. Seventeen papers reported the results of 18 different series including 5,892 cases with an average median follow-up of 92.1 months.

Results

Eight studies investigated DFS unadjusted for other variables in 2,285 cases. The relative hazard estimates ranged from 0.85 – 3.03 with a combined random effects estimate of 1.66 (95%CI 1.25 – 2.22). The effect of BCL2 on DFS adjusted for other prognostic factors was reported in 11 studies and the pooled random effects hazard ratio estimate was 1.58 (95%CI 1.29–1.94). OS was investigated unadjusted for other variables in eight studies incorporating 3,910 cases. The hazard estimates ranged from 0.99–4.31 with a pooled estimate of risk of 1.64 (95%CI 1.36–2.0). OS adjusted for other parameters was evaluated in nine series comprising 3,624 cases and the estimates for these studies ranged from 1.10 to 2.49 with a pooled estimate of 1.37 (95%CI 1.19–1.58).

Conclusion

The meta-analysis strongly supports the prognostic role of BCL2 as assessed by immunohistochemistry in breast cancer and shows that this effect is independent of lymph node status, tumour size and tumour grade as well as a range of other biological variables on multi-variate analysis. Large prospective studies are now needed to establish the clinical utility of BCL2 as an independent prognostic marker.

Effects of common germline genetic variation in cell cycle control genes on breast cancer survival: results from a population-based cohort

INTRODUCTION

Somatic alterations have been shown to correlate with breast cancer prognosis and survival, but less is known about the effects of common inherited genetic variation. Of particular interest are genes involved in cell cycle pathways, which regulate cell division.

METHODS

We examined associations between common germline genetic variation in 13 genes involved in cell cycle control (CCND1, CCND2, CCND3, CCNE1, CDK2 [p33], CDK4, CDK6, CDKN1A [p21, Cip1], CDKN1B [p27, Kip1], CDKN2A [p16], CDKN2B [p15], CDKN2C [p18], and CDKN2D [p19]) and survival among women diagnosed with invasive breast cancer participating in the SEARCH (Studies of Epidemiology and Risk factors in Cancer Heredity) breast cancer study. DNA from up to 4,470 women was genotyped for 85 polymorphisms that tag the known common polymorphisms (minor allele frequency > 0.05) in the genes. The genotypes of each polymorphism were tested for association with survival using Cox regression analysis.

RESULTS

The rare allele of the tagging single nucleotide polymorphism (SNP) rs2479717 is associated with an increased risk of death (hazard ratio = 1.26 per rare allele carried, 95% confidence interval: 1.12 to 1.42; P = 0.0001), which was not attenuated after adjusting for tumour stage, grade, and treatment. This SNP is part of a large linkage disequilibrium block, which contains CCND3, BYSL, TRFP, USP49, C6ofr49, FRS3, and PGC. We evaluated the association of survival and somatic expression of these genes in breast tumours using expression microarray data from seven published datasets. Elevated expression of the C6orf49 transcript was associated with breast cancer survival, adding biological interest to the finding.

CONCLUSION

It is possible that CCND3 rs2479717, or another variant it tags, is associated with prognosis after a diagnosis of breast cancer. Further study is required to validate this finding.

A gene expression signature that can predict the recurrence of tamoxifen-treated primary breast cancer

PURPOSE

The identification of a molecular signature predicting the relapse of tamoxifen-treated primary breast cancers should help the therapeutic management of estrogen receptor-positive cancers.

EXPERIMENTAL DESIGN

A series of 132 primary tumors from patients who received adjuvant tamoxifen were analyzed for expression profiles at the whole-genome level by 70-mer oligonucleotide microarrays. A supervised analysis was done to identify an expression signature.

RESULTS

We defined a 36-gene signature that correctly classified 78% of patients with relapse and 80% of relapse-free patients (79% accuracy). Using 23 independent tumors, we confirmed the accuracy of the signature (78%) whose relevance was further shown by using published microarray data from 60 tamoxifen-treated patients (63% accuracy). Univariate analysis using the validation set of 83 tumors showed that the 36-gene classifier is more efficient in predicting disease-free survival than the traditional histopathologic prognostic factors and is as effective as the Nottingham Prognostic Index or the "Adjuvant!" software. Multivariate analysis showed that the molecular signature is the only independent prognostic factor. A comparison with several already published signatures demonstrated that the 36-gene signature is among the best to classify tumors from both training and validation sets. Kaplan-Meier analyses emphasized its prognostic power both on the whole cohort of patients and on a subgroup with an intermediate risk of recurrence as defined by the St. Gallen criteria.

CONCLUSION

This study identifies a molecular signature specifying a subgroup of patients who do not gain benefits from tamoxifen treatment. These patients may therefore be eligible for alternative endocrine therapies and/or chemotherapy.

High-resolution aCGH and expression profiling identifies a novel genomic subtype of ER negative breast cancer

High resolution array-CGH and expression profiling identifies a novel genomic subtype of ER negative breast cancer, and provides a genome-wide list of common copy number alterations associated with aberrant expression and poor prognosis.

Background

The characterization of copy number alteration patterns in breast cancer requires high-resolution genome-wide profiling of a large panel of tumor specimens. To date, most genome-wide array comparative genomic hybridization studies have used tumor panels of relatively large tumor size and high Nottingham Prognostic Index (NPI) that are not as representative of breast cancer demographics.

Results

We performed an oligo-array-based high-resolution analysis of copy number alterations in 171 primary breast tumors of relatively small size and low NPI, which was therefore more representative of breast cancer demographics. Hierarchical clustering over the common regions of alteration identified a novel subtype of high-grade estrogen receptor (ER)-negative breast cancer, characterized by a low genomic instability index. We were able to validate the existence of this genomic subtype in one external breast cancer cohort. Using matched array expression data we also identified the genomic regions showing the strongest coordinate expression changes ('hotspots'). We show that several of these hotspots are located in the phosphatome, kinome and chromatinome, and harbor members of the 122-breast cancer CAN-list. Furthermore, we identify frequently amplified hotspots on 8q22.3 (EDD1, WDSOF1), 8q24.11-13 (THRAP6, DCC1, SQLE, SPG8) and 11q14.1 (NDUFC2, ALG8, USP35) associated with significantly worse prognosis. Amplification of any of these regions identified 37 samples with significantly worse overall survival (hazard ratio (HR) = 2.3 (1.3-1.4) p = 0.003) and time to distant metastasis (HR = 2.6 (1.4-5.1) p = 0.004) independently of NPI.

Conclusion

We present strong evidence for the existence of a novel subtype of high-grade ER-negative tumors that is characterized by a low genomic instability index. We also provide a genome-wide list of common copy number alteration regions in breast cancer that show strong coordinate aberrant expression, and further identify novel frequently amplified regions that correlate with poor prognosis. Many of the genes associated with these regions represent likely novel oncogenes or tumor suppressors.

Proliferation markers and survival in early breast cancer: a systematic review and meta-analysis of 85 studies in 32,825 patients

We have performed a systematic review and meta-analysis of proliferation markers (Ki-67, mitotic index (MI), proliferating cell nuclear antigen (PCNA) and thymidine or bromodeoxyuridine labelling index (LI)) with respect to survival in early breast cancer. Eighty-five studies involving 32,825 patients were analysed. Ki-67 (43 studies, 15,790 patients), MI (20 studies, 7021 patients), and LI (11 studies, 7337 patients) were associated with significantly shorter overall and disease free survival, using results from univariate and multivariate analyses from the individual studies. PCNA (11 studies, 2677 patients) was associated with shorter overall survival by multivariate analysis only, because of lack of data. There was some evidence for publication bias, but all markers remained significant after allowing for this. Ki-67, MI, PCNA and LI are associated with worse survival outcomes in early breast cancer. However, whether these proliferation markers provide additional prognostic information to commonly used prognostic indices remains unclear.

Cell cycle correlated genes dictate the prognostic power of breast cancer gene lists

Background

Numerous gene lists or "classifiers" have been derived from global gene expression data that assign breast cancers to good and poor prognosis groups. A remarkable feature of these molecular signatures is that they have few genes in common, prompting speculation that they may use distinct genes to measure the same pathophysiological process(es), such as proliferation. However, this supposition has not been rigorously tested. If gene-based classifiers function by measuring a minimal number of cellular processes, we hypothesized that the informative genes for these processes could be identified and the data sets could be adjusted for the predictive contributions of those genes. Such adjustment would then attenuate the predictive function of any signature measuring that same process.

Results

We tested this hypothesis directly using a novel iterative-subtractive approach. We evaluated five gene expression data sets that sample a broad range of breast cancer subtypes. In all data sets, the dominant cluster capable of predicting metastasis was heavily populated by genes that fluctuate in concert with the cell cycle. When six well-characterized classifiers were examined, all contained a higher than expected proportion of genes that correlate with this cluster. Furthermore, when the data sets were globally adjusted for the cell cycle cluster, each classifier lost its ability to assign tumors to appropriate high and low risk groups. In contrast, adjusting for other predictive gene clusters did not impact their performance.

Conclusion

These data indicate that the discriminative ability of breast cancer classifiers is dependent upon genes that correlate with cell cycle progression.

Gene expression profiling in primary breast cancer distinguishes patients developing local recurrence after breast-conservation surgery, with or without postoperative radiotherapy

Introduction

Some patients with breast cancer develop local recurrence after breast-conservation surgery despite postoperative radiotherapy, whereas others remain free of local recurrence even in the absence of radiotherapy. As clinical parameters are insufficient for identifying these two groups of patients, we investigated whether gene expression profiling would add further information.

Methods

We performed gene expression analysis (oligonucleotide arrays, 26,824 reporters) on 143 patients with lymph node-negative disease and tumor-free margins. A support vector machine was employed to build classifiers using leave-one-out cross-validation.

Results

Within the estrogen receptor-positive (ER⁺) subgroup, the gene expression profile clearly distinguished patients with local recurrence after radiotherapy (n = 20) from those without local recurrence (n = 80 with or without radiotherapy). The receiver operating characteristic (ROC) area was 0.91, and 5,237 of 26,824 reporters had a P value of less than 0.001 (false discovery rate = 0.005). This gene expression profile provides substantially added value to conventional clinical markers (for example, age, histological grade, and tumor size) in predicting local recurrence despite radiotherapy. Within the ER^- subgroup, a weaker, but still significant, signal was found (ROC area = 0.74). The ROC area for distinguishing patients who develop local recurrence from those who remain local recurrence-free in the absence of radiotherapy was 0.66 (combined ER⁺/ER^-).

Conclusion

A highly distinct gene expression profile for patients developing local recurrence after breast-conservation surgery despite radiotherapy has been identified. If verified in further studies, this profile might be a most important tool in the decision making for surgery and adjuvant therapy.

A gene sets approach for identifying prognostic gene signatures for outcome prediction

Background

Gene expression profiling is a promising approach to better estimate patient prognosis; however, there are still unresolved problems, including little overlap among similarly developed gene sets and poor performance of a developed gene set in other datasets.

Results

We applied a gene sets approach to develop a prognostic gene set from multiple gene expression datasets. By analyzing 12 independent breast cancer gene expression datasets comprising 1,756 tissues with 2,411 pre-defined gene sets including gene ontology categories and pathways, we found many gene sets that were prognostic in most of the analyzed datasets. Those prognostic gene sets were related to biological processes such as cell cycle and proliferation and had additional prognostic values over conventional clinical parameters such as tumor grade, lymph node status, estrogen receptor (ER) status, and tumor size. We then estimated the prediction accuracy of each gene set by performing external validation using six large datasets and identified a gene set with an average prediction accuracy of 67.55%.

Conclusion

A gene sets approach is an effective method to develop prognostic gene sets to predict patient outcome and to understand the underlying biology of the developed gene set. Using the gene sets approach we identified many prognostic gene sets in breast cancer.

Quantitative expression profiling of highly degraded RNA from formalin-fixed, paraffin-embedded breast tumor biopsies by oligonucleotide microarrays

Microarray-based gene expression profiling is well suited for parallel quantitative analysis of large numbers of RNAs, but its application to cancer biopsies, particularly formalin-fixed, paraffin-embedded (FFPE) archived tissues, is limited by the poor quality of the RNA recovered. This represents a serious drawback, as FFPE tumor tissue banks are available with clinical and prognostic annotations, which could be exploited for molecular profiling studies, provided that reliable analytical technologies are found. We applied and evaluated here a microarray-based cDNA-mediated annealing, selection, extension and ligation (DASL) assay for analysis of 502 mRNAs in highly degraded total RNA extracted from cultured cells or FFPE breast cancer (MT) biopsies. The study included quantitative and qualitative comparison of data obtained by analysis of the same RNAs with genome-wide oligonucleotide microarrays vs DASL arrays and, by DASL, before and after extensive in vitro RNA fragmentation. The DASL-based expression profiling assay applied to RNA extracted from MCF-7 cells, before or after 24 h stimulation with a mitogenic dose of 17beta-estradiol, consistently allowed to detect hormone-induced gene expression changes following extensive RNA degradation in vitro. Comparable results where obtained with tumor RNA extracted from FFPE MT biopsies (6 to 19 years old). The method proved itself sensitive, reproducible and accurate, when compared to results obtained by microarray analysis of RNA extracted from snap-frozen tissue of the same tumor.

An immune response gene expression module identifies a good prognosis subtype in estrogen receptor negative breast cancer

A feature selection method was used in an analysis of three major microarray expression datasets to identify molecular subclasses and prognostic markers in estrogen receptor-negative breast cancer, showing that it is a heterogeneous disease with at least four main subtypes.

Background

Estrogen receptor (ER)-negative breast cancer specimens are predominantly of high grade, have frequent p53 mutations, and are broadly divided into HER2-positive and basal subtypes. Although ER-negative disease has overall worse prognosis than does ER-positive breast cancer, not all ER-negative breast cancer patients have poor clinical outcome. Reliable identification of ER-negative tumors that have a good prognosis is not yet possible.

Results

We apply a recently proposed feature selection method in an integrative analysis of three major microarray expression datasets to identify molecular subclasses and prognostic markers in ER-negative breast cancer. We find a subclass of basal tumors, characterized by over-expression of immune response genes, which has a better prognosis than the rest of ER-negative breast cancers. Moreover, we show that, in contrast to ER-positive tumours, the majority of prognostic markers in ER-negative breast cancer are over-expressed in the good prognosis group and are associated with activation of complement and immune response pathways. Specifically, we identify an immune response related seven-gene module and show that downregulation of this module confers greater risk for distant metastasis (hazard ratio 2.02, 95% confidence interval 1.2-3.4; P = 0.009), independent of lymph node status and lymphocytic infiltration. Furthermore, we validate the immune response module using two additional independent datasets.

Conclusion

We show that ER-negative basal breast cancer is a heterogeneous disease with at least four main subtypes. Furthermore, we show that the heterogeneity in clinical outcome of ER-negative breast cancer is related to the variability in expression levels of complement and immune response pathway genes, independent of lymphocytic infiltration.

Merging microarray data from separate breast cancer studies provides a robust prognostic test

Background

There is an urgent need for new prognostic markers of breast cancer metastases to ensure that newly diagnosed patients receive appropriate therapy. Recent studies have demonstrated the potential value of gene expression signatures in assessing the risk of developing distant metastases. However, due to the small sample sizes of individual studies, the overlap among signatures is almost zero and their predictive power is often limited. Integrating microarray data from multiple studies in order to increase sample size is therefore a promising approach to the development of more robust prognostic tests.

Results

In this study, by using a highly stable data aggregation procedure based on expression comparisons, we have integrated three independent microarray gene expression data sets for breast cancer and identified a structured prognostic signature consisting of 112 genes organized into 80 pair-wise expression comparisons. A classical likelihood ratio test based on these comparisons, essentially weighted voting, achieves 88.6% sensitivity and 54.6% specificity in an independent external test set of 154 samples. The test is highly informative in assessing the risk of developing distant metastases within five years (hazard ratio 9.3 with 95% CI 2.9–29.9).

Conclusion

Rank-based features provide a stable way to integrate patient data from separate microarray studies due to invariance to data normalization, and such features can be combined into a useful predictor of distant metastases in breast cancer within a statistical modeling framework which begins to capture gene-gene interactions. Upon further confirmation on large-scale independent data, such prognostic signatures and tests could provide a powerful tool to guide adjuvant systemic treatment that could greatly reduce the cost of breast cancer treatment, both in terms of toxic side effects and health care expenditures.

Building pathway clusters from Random Forests classification using class votes

Background

Recent years have seen the development of various pathway-based methods for the analysis of microarray gene expression data. These approaches have the potential to bring biological insights into microarray studies. A variety of methods have been proposed to construct networks using gene expression data. Because individual pathways do not act in isolation, it is important to understand how different pathways coordinate to perform cellular functions. However, there are no published methods describing how to build pathway clusters that are closely related to traits of interest.

Results

We propose to build pathway clusters from pathway-based classification methods. The proposed methods allow researchers to identify clusters of pathways sharing similar functions. These pathways may or may not share genes. As an illustration, our approach is applied to three human breast cancer microarray data sets. We found that our methods yielded consistent and interpretable results for these three data sets. We further investigated one of the pathway clusters found using PubMatrix. We found that informative genes in the pathway clusters do have more publications with keywords, like estrogen receptor, compared with informative genes in other top pathways. In addition, using the shortest path analysis in GeneGo's MetaCore and Human Protein Reference Database, we were able to identify the links which connect the pathways without shared genes within the pathway cluster.

Conclusion

Our proposed pathway clustering methods allow bioinformaticians and biologists to investigate how informative genes within pathways are related to each other and understand possible crosstalk between pathways in a cluster. Therefore, building pathway clusters may lead to a better understanding of molecular mechanisms affecting a trait of interest, and help generate further biological hypotheses from gene expression data.

ANCCA, an estrogen-regulated AAA+ ATPase coactivator for ERalpha, is required for coregulator occupancy and chromatin modification

AAA+ proteins play crucial roles in diverse biological processes via their ATPase-driven remodeling of macromolecular complexes. Here we report our identification of an evolutionarily conserved AAA+ protein, ANCCA/pro2000, endowed with a bromodomain that is strongly induced by estrogen in human breast cancer cells and is a direct target of protooncogene ACTR/AIB1/SRC-3. We found that ANCCA associates directly with estrogen-bound estrogen receptor (ER) alpha and ACTR. It is selectively recruited, upon estrogen stimulation, to a subset of ERalpha target genes including cyclin D1, c-myc, and E2F1 and is required for their estrogen-induced expression as well as breast cancer cell proliferation. Further studies indicate that ANCCA binds and hydrolyzes ATP and is critical for recruitment of coregulator CBP and histone hyperacetylation at the ER target chromatin. Moreover, mutations at the ATP binding motifs rendered ANCCA defective as a coactivator in mediating estrogen induction of gene expression. Together, our findings reveal an unexpected layer of regulatory mechanism in hormone signaling mediated by ANCCA and suggest that hormone-induced assembly of transcriptional coregulator complexes at chromatin is a process facilitated by AAA+ ATPase proteins.

Module-based outcome prediction using breast cancer compendia

Background

The availability of large collections of microarray datasets (compendia), or knowledge about grouping of genes into pathways (gene sets), is typically not exploited when training predictors of disease outcome. These can be useful since a compendium increases the number of samples, while gene sets reduce the size of the feature space. This should be favorable from a machine learning perspective and result in more robust predictors.

Methodology

We extracted modules of regulated genes from gene sets, and compendia. Through supervised analysis, we constructed predictors which employ modules predictive of breast cancer outcome. To validate these predictors we applied them to independent data, from the same institution (intra-dataset), and other institutions (inter-dataset).

Conclusions

We show that modules derived from single breast cancer datasets achieve better performance on the validation data compared to gene-based predictors. We also show that there is a trend in compendium specificity and predictive performance: modules derived from a single breast cancer dataset, and a breast cancer specific compendium perform better compared to those derived from a human cancer compendium. Additionally, the module-based predictor provides a much richer insight into the underlying biology. Frequently selected gene sets are associated with processes such as cell cycle, E2F regulation, DNA damage response, proteasome and glycolysis. We analyzed two modules related to cell cycle, and the OCT1 transcription factor, respectively. On an individual basis, these modules provide a significant separation in survival subgroups on the training and independent validation data.

Molecular heterogeneity of breast carcinomas and the cancer stem cell hypothesis

Human breast cancers are heterogeneous, both in their pathology and in their molecular profiles. This suggests the hypothesis that breast cancers can initiate in different cell types, either breast epithelial stem cells or their progeny (transit amplifying cells or committed differentiated cells). In this respect, breast cancer could be viewed as being similar to haematological malignancies for which an analogous model has been proposed. Drawing such parallels might help to unravel the molecular nature of the initiating events in breast cancer and might have substantial clinical implications.

Integrated analysis of independent gene expression microarray datasets improves the predictability of breast cancer outcome

BACKGROUND

Gene expression profiles based on microarray data have been suggested by many studies as potential molecular prognostic indexes of breast cancer. However, due to the confounding effect of clinical background, independent studies often obtained inconsistent results. The current study investigated the possibility to improve the quality and generality of expression profiles by integrated analysis of multiple datasets. Profiles of recurrence outcome were derived from two independent datasets and validated by a third dataset.

RESULTS

The clinical background of patients significantly influenced the content and performance of expression profiles when the training samples were unbalanced. The integrated profiling of two independent datasets lead to higher classification accuracy (71.11% vs. 70.59%) and larger ROC curve area (0.789 vs. 0.767) of the testing samples. Cell cycle, especially M phase mitosis, was significantly overrepresented by the 60-gene profile obtained from integrated analysis (p < 0.0001). This profiles significantly differentiated poor and good prognosis in a third patient cohort (p = 0.003). Simulation procedures demonstrated that the change of profile specificity had more instant influence on the performance of expression profiles than the change of profile sensitivity.

CONCLUSION

The current study confirmed that the gene expression profile generated by integrated analysis of multiple datasets achieved better prediction of breast cancer recurrence. However, the content and performance of profiles was confounded by clinical background of training patients. In future studies, prognostic profile applicable to the general population should be derived from more diversified and balanced patient cohorts in larger scale.

The phenotypic spectrum of basal-like breast cancers: a critical appraisal

There are 2 well-recognized cell populations lining the mammary duct system: the epithelial cells lining the lumen and the myoepithelial cells surrounding them. The mammary stem cell, a putative third cell type, has not yet been well characterized. It is not established whether the putative stem cell expresses the full complement, a subset, or none of the markers of normal epithelial and/or myoepithelial cells. However, it is likely that they would have distinctive markers of their own; whether these are retained or lost in their neoplastic progeny is unknown. All 3 cell types may theoretically undergo malignant transformation. Until recently, however, nearly all attention has been focused on carcinomas of epithelial derivation/differentiation. The advent of oligonucleotide and cDNA microarrays has facilitated gene expression profiling of breast cancers, revealing molecular subclasses that may be prognostically relevant. One such subclass, the basal-like breast carcinomas, has been found in numerous independent datasets to be associated with a comparatively worse overall and disease-free survival. These cancers show expression of molecules characteristic of the normal myoepithelial cell, such as basal cytokeratins, and reduced expression of estrogen receptor-related and Erb-B2-related genes and proteins. The classifier genes that formed the basis for the delineation of basal-like carcinomas were derived from datasets that were composed predominantly of ductal type cancers. Therefore, the clinical significance of a basal-like gene expression or immunohistochemical profile in the other breast cancer subtypes is presently unknown. Herein, we evaluate in detail the current state of knowledge on the pathologic features of breast carcinomas classified as basal-like by immunohistochemical and/or gene expression profiling criteria, with an emphasis on their full phenotypic spectrum and also previously underemphasized areas of heterogeneity and ambiguity where present. There seems to be a phenotypic and biologic spectrum of basal-like or myoepithelial-type carcinomas, just as there is a wide range among tumors of luminal epithelial derivation/differentiation. It is critical to promote lucid morphologic definitions of the molecular subtypes, if this information is intended for use in targeted therapies and patient management.

Taking gene-expression profiling to the clinic: when will molecular signatures become relevant to patient care?

The advent of microarray technology has enabled scientists to simultaneously investigate the expression of thousands of genes. Gene-expression profiling studies have provided a molecular classification of breast cancer into clinically relevant subtypes, new tools to predict disease recurrence and response to different treatments, and new insights into various oncogenic pathways and the process of metastatic progression. Here we describe the state of the art of gene-expression studies in breast cancer, and consider both their current limitations and future promises. We also discuss the potential of molecular signatures to have an impact on individual breast cancer patient management, and ultimately to accelerate the transition between empirical and tailored medicine.

Elucidating the altered transcriptional programs in breast cancer using independent component analysis

The quantity of mRNA transcripts in a cell is determined by a complex interplay of cooperative and counteracting biological processes. Independent Component Analysis (ICA) is one of a few number of unsupervised algorithms that have been applied to microarray gene expression data in an attempt to understand phenotype differences in terms of changes in the activation/inhibition patterns of biological pathways. While the ICA model has been shown to outperform other linear representations of the data such as Principal Components Analysis (PCA), a validation using explicit pathway and regulatory element information has not yet been performed. We apply a range of popular ICA algorithms to six of the largest microarray cancer datasets and use pathway-knowledge and regulatory-element databases for validation. We show that ICA outperforms PCA and clustering-based methods in that ICA components map closer to known cancer-related pathways, regulatory modules, and cancer phenotypes. Furthermore, we identify cancer signalling and oncogenic pathways and regulatory modules that play a prominent role in breast cancer and relate the differential activation patterns of these to breast cancer phenotypes. Importantly, we find novel associations linking immune response and epithelial–mesenchymal transition pathways with estrogen receptor status and histological grade, respectively. In addition, we find associations linking the activity levels of biological pathways and transcription factors (NF1 and NFAT) with clinical outcome in breast cancer. ICA provides a framework for a more biologically relevant interpretation of genomewide transcriptomic data. Adopting ICA as the analysis tool of choice will help understand the phenotype–pathway relationship and thus help elucidate the molecular taxonomy of heterogeneous cancers and of other complex genetic diseases.

The amount of a given transcript or protein in a cell is determined by a balance of expression and repression in a complex network of biological processes. This delicate balance is compromised in complex genetic diseases such as cancer by alterations in the activation patterns of functionally important biological processes known as pathways. Over the last years, a large number of microarray experiments profiling the expression levels of more than 20,000 human genes in hundreds of tumor samples have shown that most cancer types are heterogeneous diseases, each characterized by many different expression subtypes. The biological and clinical goal is to explain the observed tumor and clinical heterogeneity in terms of specific patterns of altered pathways. The bioinformatic challenge is therefore to devise mathematical tools that explicitly attempt to infer these altered pathways. To this end, we applied a signal processing tool in a meta-analysis of breast cancer, encompassing more than 800 tumor specimens derived from four different patient cohorts, and showed that this algorithm significantly outperforms popular standard bioinformatics tools in identifying altered pathways underlying breast cancer. These results show that the same tool could be applied to other complex human genetic diseases to better elucidate the underlying altered pathways.