A consensus prognostic gene expression classifier for ER positive breast cancer

Quick Automatic Synthesis of Solvent-Free 16α-[18F] Fluoroestradiol: Comparison of Kryptofix 222 and Tetrabutylammonium Bicarbonate

Estrogen receptor (ER) expression level of human breast cancer often reflects the stage of disease and is usually monitored by immunohistochemical staining in vitro. The preferable non-invasive and real-time diagnosis in vivo is more accessible by PET scan using 16α-[¹⁸F]FES. The objective of this study was to develop a quick automatic method for synthesis of solvent-free 16α-[¹⁸F]FES using a CFN-MPS-200 synthesis system and compare the catalytic efficiency of two phase transfer catalysts, Kryptofix 222/K₂CO₃ (K222/K₂CO₃) and tetrabutylammonium hydrogen carbonate (TBA·HCO₃). In this method, phase transfer catalysts K222/K₂CO₃ and TBA·HCO₃ were used, respectively. The intermediate products were both hydrolyzed with hydrochloric acid and neutralized with sodium bicarbonate. The crude product was purified with semi-preparative HPLC, and the solvent was removed by rotary evaporation. The effects of radiofluorination temperature and time on the synthesis were also investigated. Radiochemical purity of solvent-free product was above 99% and the decay-corrected radiochemical yield of 16α-[¹⁸F]FES was obtained in 48.7 ± 0.95% (catalyzed by K222/K₂CO₃, n = 4) and 46.7 ± 0.77% (catalyzed by TBA·HCO₃, n = 4, respectively). The solvent-free 16α-[¹⁸F]FES was studied in clinically diagnosed breast cancer patients, and FES-PET results were compared with pathology diagnosis results to validate the diagnosis value of 16α-[¹⁸F]FES. The new method was more reliable, efficient, and time-saving. There was no significant difference in catalytic activity between K222/K₂CO₃ and TBA·HCO₃.

EPISCORE: cell type deconvolution of bulk tissue DNA methylomes from single-cell RNA-Seq data

Cell type heterogeneity presents a challenge to the interpretation of epigenome data, compounded by the difficulty in generating reliable single-cell DNA methylomes for large numbers of cells and samples. We present EPISCORE, a computational algorithm that performs virtual microdissection of bulk tissue DNA methylation data at single cell-type resolution for any solid tissue. EPISCORE applies a probabilistic epigenetic model of gene regulation to a single-cell RNA-seq tissue atlas to generate a tissue-specific DNA methylation reference matrix, allowing quantification of cell-type proportions and cell-type-specific differential methylation signals in bulk tissue data. We validate EPISCORE in multiple epigenome studies and tissue types.

Squalene epoxidase as a promising metabolic target in cancer treatment

Oncogenic alteration of the cholesterol synthesis pathway is a recognized mechanism of metabolic adaptation. In the present review, we focus on squalene epoxidase (SE), one of the two rate-limiting enzymes in cholesterol synthesis, retracing its history since its discovery as an antimycotic target to its description as an emerging metabolic oncogene by amplification with clinical relevance in cancer. We review the published literature assessing the association between SE over-expression and poor prognosis in this disease. We assess the works demonstrating how SE promotes tumor cell proliferation and migration, and displaying evidence of cancer cell demise in presence of human SE inhibitors in in vitro and in vivo models. Taken together, robust scientific evidence has by now accumulated pointing out SE as a promising novel therapeutic target in cancer treatment.

Subtype classification and heterogeneous prognosis model construction in precision medicine

Common diseases including cancer are heterogeneous. It is important to discover disease subtypes and identify both shared and unique risk factors for different disease subtypes. The advent of high-throughput technologies enriches the data to achieve this goal, if necessary statistical methods are developed. Existing methods can accommodate both heterogeneity identification and variable selection under parametric models, but for survival analysis, the commonly used Cox model is semiparametric. Although finite-mixture Cox model has been proposed to address heterogeneity in survival analysis, variable selection has not been incorporated into such semiparametric models. Using regularization regression, we propose a variable selection method for the finite-mixture Cox model and select important, subtype-specific risk factors from high-dimensional predictors. Our estimators have oracle properties with proper choices of penalty parameters under the regularization regression. An expectation-maximization algorithm is developed for numerical calculation. Simulations demonstrate that our proposed method performs well in revealing the heterogeneity and selecting important risk factors for each subtype, and its performance is compared to alternatives with other regularizers. Finally, we apply our method to analyze a gene expression dataset for ovarian cancer DNA repair pathways. Based on our selected risk factors, the prognosis model accounting for heterogeneity consistently improves the prediction for the survival probability in both training and test datasets.

Prolactin-induced protein as a potential therapy response marker of adjuvant chemotherapy in breast cancer patients

Many studies are dedicated to exploring the molecular mechanisms of chemotherapy-resistance in breast cancer (BC). Some of them are focused on searching for candidate genes responsible for this process. The aim of this study was typing the candidate genes associated with the response to standard chemotherapy in the case of invasive ductal carcinoma. Frozen material from 28 biopsies obtained from IDC patients with different responses to chemotherapy were examined using gene expression microarray, Real-Time PCR (RT-PCR) and Western blot (WB). Based on the microarray results, further analysis of candidate gene expression was evaluated in 120 IDC cases by RT-PCR and in 224 IDC cases by immunohistochemistry (IHC). The results were correlated with clinical outcome and molecular subtype of the BC. Gene expression microarray revealed Prolactin-Induced Peptide (PIP) as a single gene differentially expressed in BC therapy responder or non-responder patients (p <0.05). The level of PIP expression was significantly higher in the BC therapy responder group than in the non-responder group at mRNA (p=0.0092) and protein level (p=0.0256). Expression of PIP mRNA was the highest in estrogen receptor positive (ER+) BC cases (p=0.0254) and it was the lowest in triple negative breast cancer (TNBC) (p=0.0336). Higher PIP mRNA expression was characterized by significantly longer disease free survival (DFS, p=0.0093), as well as metastasis free survival (MFS, p=0.0144). Additionally, PIP mRNA and PIP protein expression levels were significantly higher in luminal A than in other molecular subtypes and TNBC. Moreover significantly higher PIP expression was observed in G1, G2 vs. G3 cases (p=0.0027 and p=0.0013, respectively). Microarray analysis characterized PIP gene as a candidate for BC standard chemotherapy response marker. Analysis of clinical data suggests that PIP may be a good prognostic and predictive marker in IDC patients. Higher levels of PIP were related to longer DFS and MFS but not with OS.

Low Concordance between Gene Expression Signatures in ER Positive HER2 Negative Breast Carcinoma Could Impair Their Clinical Application

BACKGROUND

Numerous prognostic gene expression signatures have been recently described. Among the signatures there is variation in the constituent genes that are utilized. We aim to evaluate prognostic concordance among eight gene expression signatures, on a large dataset of ER positive HER2 negative breast cancers.

METHODS

We analysed the performance of eight gene expression signatures on six different datasets of ER+ HER2- breast cancers. Survival analyses were performed using the Kaplan-Meier estimate of survival function. We assessed discrimination and concordance between the 8 signatures on survival and recurrence rates The Nottingham Prognostic Index (NPI) was used to to stratify the risk of recurrence/death.

RESULTS

The discrimination ability of the whole signatures, showed fair discrimination performances, with AUC ranging from 0.64 (95%CI 0.55-0.73 for the 76-genes signatures, to 0.72 (95%CI 0.64-0.8) for the Molecular Prognosis Index T17. Low concordance was found in predicting events in the intermediate and high-risk group, as defined by the NPI. Low risk group was the only subgroup with a good signatures concordance.

CONCLUSION

Genomic signatures may be a good option to predict prognosis as most of them perform well at the population level. They exhibit, however, a high degree of discordance in the intermediate and high-risk groups. The major benefit that we could expect from gene expression signatures is the standardization of proliferation assessment.

Expression of a-Tocopherol-Associated protein (TAP) is associated with clinical outcome in breast cancer patients

BACKGROUND

The role of vitamin E in breast cancer prevention and treatment has been widely investigated, and the different tocopherols that comprise this nutrient have been shown to have divergent associations with cancer outcome. Our previous studies have shown that α-Tocopherol-associated protein (TAP), a vitamin E binding protein, may function as a tumor suppressor-like factor in breast carcinogenesis. The current study addresses the association of TAP expression with breast cancer clinical outcomes.

METHODS

Immunohistochemical stain for TAP was applied to a tissue microarray from a breast cancer cohort consisting of 271 patients with a median follow-up time of 5.2 years. The expression of TAP in tumor cells was compared with patient's clinical outcome at 5 years after diagnosis. The potential role of TAP in predicting outcome was also assessed in clinically relevant subsets of the cohort. In addition, we compared TAP expression and Oncotype DX scores in an independent breast cancer cohort consisting of 71 cases.

RESULTS

We demonstrate that the expression of TAP was differentially expressed within the breast cancer cohort, and that ER+/PR ± tumors were more likely to exhibit TAP expression. TAP expression was associated with an overall lower recurrence rate and a better 5-year survival rate. This association was primarily in patients with ER+ tumors; exploratory analysis showed that this association was strongest in patients with node-positive tumors and was independent of stage and treatment with chemotherapy. TAP expression in ER/PR negative or triple negative tumors had no association with clinical outcome. In addition, we did not observe an association between TAP expression and Oncotype DX recurrence score.

CONCLUSIONS

The significant positive association we found for α-Tocopherol-associated protein with outcome in breast cancer may help to better define and explain studies addressing α-tocopherol's association with cancer risk and outcome. Additionally, further studies to validate and extend these findings may allow TAP to serve as a breast-specific prognostic marker in breast cancer patients, especially in those patients with ER+ tumors.

Denoising perturbation signatures reveal an actionable AKT-signaling gene module underlying a poor clinical outcome in endocrine-treated ER+ breast cancer

Background

Databases of perturbation gene expression signatures and drug sensitivity provide a powerful framework to develop personalized medicine approaches, by helping to identify actionable genomic markers and subgroups of patients who may benefit from targeted treatments.

Results

Here we use a perturbation expression signature database encompassing perturbations of over 90 cancer genes, in combination with a large breast cancer expression dataset and a novel statistical denoising algorithm, to help discern cancer perturbations driving most of the variation in breast cancer gene expression. Clustering estrogen receptor positive cancers over the perturbation activity scores recapitulates known luminal subtypes. Analysis of individual activity scores enables identification of a novel cancer subtype, defined by a 31-gene AKT-signaling module. Specifically, we show that activation of this module correlates with a poor prognosis in over 900 endocrine-treated breast cancers, a result we validate in two independent cohorts. Importantly, breast cancer cell lines with high activity of the module respond preferentially to PI3K/AKT/mTOR inhibitors, a result we also validate in two independent datasets. We find that at least 34 % of the downregulated AKT module genes are either mediators of apoptosis or have tumor suppressor functions.

Conclusions

The statistical framework advocated here could be used to identify gene modules that correlate with prognosis and sensitivity to alternative treatments. We propose a randomized clinical trial to test whether the 31-gene AKT module could be used to identify estrogen receptor positive breast cancer patients who may benefit from therapy targeting the PI3K/AKT/mTOR signaling axis.

Electronic supplementary material

The online version of this article (doi:10.1186/s13059-015-0630-4) contains supplementary material, which is available to authorized users.

Evaluation of public cancer datasets and signatures identifies TP53 mutant signatures with robust prognostic and predictive value

Background

Systematic analysis of cancer gene-expression patterns using high-throughput transcriptional profiling technologies has led to the discovery and publication of hundreds of gene-expression signatures. However, few public signature values have been cross-validated over multiple studies for the prediction of cancer prognosis and chemosensitivity in the neoadjuvant setting.

Methods

To analyze the prognostic and predictive values of publicly available signatures, we have implemented a systematic method for high-throughput and efficient validation of a large number of datasets and gene-expression signatures. Using this method, we performed a meta-analysis including 351 publicly available signatures, 37,000 random signatures, and 31 breast cancer datasets. Survival analyses and pathologic responses were used to assess prediction of prognosis, chemoresponsiveness, and chemo-drug sensitivity.

Results

Among 31 breast cancer datasets and 351 public signatures, we identified 22 validation datasets, two robust prognostic signatures (BRmet50 and PMID18271932Sig33) in breast cancer and one signature (PMID20813035Sig137) specific for prognosis prediction in patients with ER-negative tumors. The 22 validation datasets demonstrated enhanced ability to distinguish cancer gene profiles from random gene profiles. Both prognostic signatures are composed of genes associated with TP53 mutations and were able to stratify the good and poor prognostic groups successfully in 82%and 68% of the 22 validation datasets, respectively. We then assessed the abilities of the two signatures to predict treatment responses of breast cancer patients treated with commonly used chemotherapeutic regimens. Both BRmet50 and PMID18271932Sig33 retrospectively identified those patients with an insensitive response to neoadjuvant chemotherapy (mean positive predictive values 85%-88%). Among those patients predicted to be treatment sensitive, distant relapse-free survival (DRFS) was improved (negative predictive values 87%-88%). BRmet50 was further shown to prospectively predict taxane-anthracycline sensitivity in patients with HER2-negative (HER2-) breast cancer.

Conclusions

We have developed and applied a high-throughput screening method for public cancer signature validation. Using this method, we identified appropriate datasets for cross-validation and two robust signatures that differentiate TP53 mutation status and have prognostic and predictive value for breast cancer patients.

Electronic supplementary material

The online version of this article (doi:10.1186/s12885-015-1102-7) contains supplementary material, which is available to authorized users.

Integrative analysis of survival-associated gene sets in breast cancer

BACKGROUND

Patient gene expression information has recently become a clinical feature used to evaluate breast cancer prognosis. The emergence of prognostic gene sets that take advantage of these data has led to a rich library of information that can be used to characterize the molecular nature of a patient's cancer. Identifying robust gene sets that are consistently predictive of a patient's clinical outcome has become one of the main challenges in the field.

METHODS

We inputted our previously established BASE algorithm with patient gene expression data and gene sets from MSigDB to develop the gene set activity score (GSAS), a metric that quantitatively assesses a gene set's activity level in a given patient. We utilized this metric, along with patient time-to-event data, to perform survival analyses to identify the gene sets that were significantly correlated with patient survival. We then performed cross-dataset analyses to identify robust prognostic gene sets and to classify patients by metastasis status. Additionally, we created a gene set network based on component gene overlap to explore the relationship between gene sets derived from MSigDB. We developed a novel gene set based on this network's topology and applied the GSAS metric to characterize its role in patient survival.

RESULTS

Using the GSAS metric, we identified 120 gene sets that were significantly associated with patient survival in all datasets tested. The gene overlap network analysis yielded a novel gene set enriched in genes shared by the robustly predictive gene sets. This gene set was highly correlated to patient survival when used alone. Most interestingly, removal of the genes in this gene set from the gene pool on MSigDB resulted in a large reduction in the number of predictive gene sets, suggesting a prominent role for these genes in breast cancer progression.

CONCLUSIONS

The GSAS metric provided a useful medium by which we systematically investigated how gene sets from MSigDB relate to breast cancer patient survival. We used this metric to identify predictive gene sets and to construct a novel gene set containing genes heavily involved in cancer progression.

Prolactin-induced protein is required for cell cycle progression in breast cancer

Prolactin-induced protein (PIP) is expressed in the majority of breast cancers and is used for the diagnostic evaluation of this disease as a characteristic biomarker; however, the molecular mechanisms of PIP function in breast cancer have remained largely unknown. In this study, we carried out a comprehensive investigation of PIP function using PIP silencing in a broad group of breast cancer cell lines, analysis of expression microarray data, proteomic analysis using mass spectrometry, and biomarker studies on breast tumors. We demonstrated that PIP is required for the progression through G1 phase, mitosis, and cytokinesis in luminal A, luminal B, and molecular apocrine breast cancer cells. In addition, PIP expression is associated with a transcriptional signature enriched with cell cycle genes and regulates key genes in this process including cyclin D1, cyclin B1, BUB1, and forkhead box M1 (FOXM1). It is notable that defects in mitotic transition and cytokinesis following PIP silencing are accompanied by an increase in aneuploidy of breast cancer cells. Importantly, we have identified novel PIP-binding partners in breast cancer and shown that PIP binds to β-tubulin and is necessary for microtubule polymerization. Furthermore, PIP interacts with actin-binding proteins including Arp2/3 and is needed for inside-out activation of integrin-β1 mediated through talin. This study suggests that PIP is required for cell cycle progression in breast cancer and provides a rationale for exploring PIP inhibition as a therapeutic approach in breast cancer that can potentially target microtubule polymerization.

ATAD2 is highly expressed in ovarian carcinomas and indicates poor prognosis

The purpose of this study was to explore the expression of ATAD2 in ovarian tumor tissue as well as its relationship with degree of malignancy. Tumor tissue from 110 cases of ovarian cancer was collected in accordance with the Declaration of Helsinki for evaluation of ATAD2 expression immunohistochemistry, quantitative PCR (qPCR) and Western blotting. The correlation between the ATAD2 expression and and the prognosis of ovarian cancer was evaluated by Cox regression model. In addition, HO-8910 and OVCAR-3 cells were transfected with two siRNAs targeting ATAD2. Cell viability was evaluated with MTT assay, and cell migration by transwell migration assay. ATAD2 was shown to be highly expressed in 65.5% (72/110) of ovarian cancer cases, both at transcriptional and protein levels. Moreover, highly expression was positively correlated with degree of malignancy. Knock-down of ATAD2 in HO-8910 and OVCAR-3 cells was found to reduce cell migration. In addition, follow-up visits of the patients demonstrated that the 5-year survival rate was lower in patients with high expression of ATAD2. Our study suggested that ovarian tumor tissue may have highly expressed ATAD2, which is associated with tumor stage, omentum-metastasis, ascites and CA-125. Increased ATAD2 may play important roles in tumor proliferation and migration. ATAD2 could serve in particular as a prognostic marker and a therapeutic target for ovarian cancer.

A comparative study of improvements Pre-filter methods bring on feature selection using microarray data

Background

Feature selection techniques have become an apparent need in biomarker discoveries with the development of microarray. However, the high dimensional nature of microarray made feature selection become time-consuming. To overcome such difficulties, filter data according to the background knowledge before applying feature selection techniques has become a hot topic in microarray analysis. Different methods may affect final results greatly, thus it is important to evaluate these pre-filter methods in a system way.

Methods

In this paper, we compared the performance of statistical-based, biological-based pre-filter methods and the combination of them on microRNA-mRNA parallel expression profiles using L1 logistic regression as feature selection techniques. Four types of data were built for both microRNA and mRNA expression profiles.

Results

Results showed that pre-filter methods could reduce the number of features greatly for both mRNA and microRNA expression datasets. The features selected after pre-filter procedures were shown to be significant in biological levels such as biology process and microRNA functions. Analyses of classification performance based on precision showed the pre-filter methods were necessary when the number of raw features was much bigger than that of samples. All the computing time was greatly shortened after pre-filter procedures.

Conclusions

With similar or better classification improvements, less but biological significant features, pre-filter-based feature selection should be taken into consideration if researchers need fast results when facing complex computing problems in bioinformatics.

Electronic supplementary material

The online version of this article (doi:10.1186/2047-2501-2-7) contains supplementary material, which is available to authorized users.

Coagulation factor VII is regulated by androgen receptor in breast cancer

Androgen receptor (AR) is widely expressed in breast cancer; however, there is limited information on the key molecular functions and gene targets of AR in this disease. In this study, gene expression data from a cohort of 52 breast cancer cell lines was analyzed to identify a network of AR co-expressed genes. A total of 300 genes, which were significantly enriched for cell cycle and metabolic functions, showed absolute correlation coefficients (|CC|) of more than 0.5 with AR expression across the dataset. In this network, a subset of 35 "AR-signature" genes were highly co-expressed with AR (|CC|>0.6) that included transcriptional regulators PATZ1, NFATC4, and SPDEF. Furthermore, gene encoding coagulation factor VII (F7) demonstrated the closest expression pattern with AR (CC=0.716) in the dataset and factor VII protein expression was significantly associated to that of AR in a cohort of 209 breast tumors. Moreover, functional studies demonstrated that AR activation results in the induction of factor VII expression at both transcript and protein levels and AR directly binds to a proximal region of F7 promoter in breast cancer cells. Importantly, AR activation in breast cancer cells induced endogenous factor VII activity to convert factor X to Xa in conjunction with tissue factor. In summary, F7 is a novel AR target gene and AR activation regulates the ectopic expression and activity of factor VII in breast cancer cells. These findings have functional implications in the pathobiology of thromboembolic events and regulation of factor VII/tissue factor signaling in breast cancer.

Prognostic significance of geminin expression levels in Ki67-high subset of estrogen receptor-positive and HER2-negative breast cancers

BACKGROUND

Indication for chemotherapy in estrogen receptor (ER)-positive and human epidermal growth factor receptor 2 (HER2)-negative breast cancers is determined on the basis of Ki67 expression level. However, since Ki67-high cancers are not necessarily sensitive to chemotherapy, identification of such patients who do not need chemotherapy is an important issue.

PATIENTS AND METHODS

We used immunohistochemical staining to examine the expression levels of ER, progesterone receptor (PgR), Ki67, and geminin, a marker of S to G2/M phases, in 80 ER-positive/HER2-negative breast cancers. The labeling indices of Ki67 and geminin were determined and cutoff values were set at 15 and 6 %, respectively.

RESULTS

Ki67 and geminin expression levels were significantly associated with nuclear grade. In the Ki67-low subset, 26 out of 28 (92.9 %) cancers were geminin low and in the Ki67-high subset, 31 out of 52 (59.6 %) were geminin high. Distant disease-free survival (DDFS) of the geminin-high subset was significantly poorer than that of the geminin-low subset (P = 0.009). In the Ki67-low subset, only one patient showed recurrence. Metastasis was detected in eight out of 31 (25.8 %) patients in the geminin-high group of the Ki67-high subset, but no recurrence was observed in the geminin-low group of the Ki67-high subset.

CONCLUSION

Geminin-high breast cancers are significantly associated with worse prognosis. Since poorer prognosis was recognized only in the geminin-high group in Ki67-high cancers, we speculate that geminin may be useful for identifying patients in the Ki67-high subset who can avoid unnecessary chemotherapy.

Breast cancer prognosis risk estimation using integrated gene expression and clinical data

BACKGROUND

Novel prognostic markers are needed so newly diagnosed breast cancer patients do not undergo any unnecessary therapy. Various microarray gene expression datasets based studies have generated gene signatures to predict the prognosis outcomes, while ignoring the large amount of information contained in established clinical markers. Nevertheless, small sample sizes in individual microarray datasets remain a bottleneck in generating robust gene signatures that show limited predictive power. The aim of this study is to achieve high classification accuracy for the good prognosis group and then achieve high classification accuracy for the poor prognosis group.

METHODS

We propose a novel algorithm called the IPRE (integrated prognosis risk estimation) algorithm. We used integrated microarray datasets from multiple studies to increase the sample sizes (∼ 2,700 samples). The IPRE algorithm consists of a virtual chromosome for the extraction of the prognostic gene signature that has 79 genes, and a multivariate logistic regression model that incorporates clinical data along with expression data to generate the risk score formula that accurately categorizes breast cancer patients into two prognosis groups.

RESULTS

The evaluation on two testing datasets showed that the IPRE algorithm achieved high classification accuracies of 82% and 87%, which was far greater than any existing algorithms.

Application of microarray in breast cancer: An overview

There are more than 1.15 million cases of breast cancer diagnosed worldwide annually. At present, only small numbers of accurate prognostic and predictive factors are used clinically for managing the patients with breast cancer. DNA microarrays have the potential to assess the expression of thousands of genes simultaneously. Recent preliminary researches indicate that gene expression profiling based on DNA microarray can offer potential and independent prognostic information in patients with newly diagnosed breast cancer. In this paper, an overview upon the applications of microarray techniques in breast cancer is presented.

A data similarity-based strategy for meta-analysis of transcriptional profiles in cancer

Background

Robust transcriptional signatures in cancer can be identified by data similarity-driven meta-analysis of gene expression profiles. An unbiased data integration and interrogation strategy has not previously been available.

Methods and Findings

We implemented and performed a large meta-analysis of breast cancer gene expression profiles from 223 datasets containing 10,581 human breast cancer samples using a novel data similarity-based approach (iterative EXALT). Cancer gene expression signatures extracted from individual datasets were clustered by data similarity and consolidated into a meta-signature with a recurrent and concordant gene expression pattern. A retrospective survival analysis was performed to evaluate the predictive power of a novel meta-signature deduced from transcriptional profiling studies of human breast cancer. Validation cohorts consisting of 6,011 breast cancer patients from 21 different breast cancer datasets and 1,110 patients with other malignancies (lung and prostate cancer) were used to test the robustness of our findings. During the iterative EXALT analysis, 633 signatures were grouped by their data similarity and formed 121 signature clusters. From the 121 signature clusters, we identified a unique meta-signature (BRmet50) based on a cluster of 11 signatures sharing a phenotype related to highly aggressive breast cancer. In patients with breast cancer, there was a significant association between BRmet50 and disease outcome, and the prognostic power of BRmet50 was independent of common clinical and pathologic covariates. Furthermore, the prognostic value of BRmet50 was not specific to breast cancer, as it also predicted survival in prostate and lung cancers.

Conclusions

We have established and implemented a novel data similarity-driven meta-analysis strategy. Using this approach, we identified a transcriptional meta-signature (BRmet50) in breast cancer, and the prognostic performance of BRmet50 was robust and applicable across a wide range of cancer-patient populations.

Lack of expression of the proteins GMPR2 and PPARα are associated with the basal phenotype and patient outcome in breast cancer

Basal-like tumours (BP) are a poor prognostic class of breast cancer but remain a biologically and clinically heterogeneous group. We have previously identified two novel genes PPARα (positive) and GMPR2 (negative) whose expression was significantly associated with BP at the transcriptome level. In this study, using a large and well-characterised series of operable invasive breast carcinomas (1,043 cases) prepared as TMAs, we assessed these targets at the protein level using immunohistochemistry and investigated associations with clinicopathological variables and patient outcome.

RESULTS

Lack of PPARα and GMPR2 protein expression was associated with BP, as defined by the expression of cytokeratin (CK) 5/6 and/or CK14, (p = 0.023, p = 0.001, respectively) or as triple-negative (ER-, PR-, HER2-) phenotype (p < 0.001 for both proteins). Positive expression of both markers was associated ER and PR positive status (p < 0.05) and with the good Nottingham Prognostic Index group (p = 0.012, p < 0.001, respectively). Univariate survival analysis showed an association between lack of expression of PPARα and GMPR2 and poor outcome in terms of shorter disease-free survival and shorter breast cancer-specific survival, respectively. However, multivariate analysis showed that these associations were not independent of other prognostic variables, namely tumour size, grade, and nodal stage. In conclusion, this study demonstrates that loss of expression of GMPR2 and PPARα is associated with BP at the protein level; indicating that they may play a role in carcinogenesis of this molecularly complex and clinically important subtype. Further studies into their relevance in further classification of BP are warranted.

Integration of clinical and gene expression data has a synergetic effect on predicting breast cancer outcome

Breast cancer outcome can be predicted using models derived from gene expression data or clinical data. Only a few studies have created a single prediction model using both gene expression and clinical data. These studies often remain inconclusive regarding an obtained improvement in prediction performance. We rigorously compare three different integration strategies (early, intermediate, and late integration) as well as classifiers employing no integration (only one data type) using five classifiers of varying complexity. We perform our analysis on a set of 295 breast cancer samples, for which gene expression data and an extensive set of clinical parameters are available as well as four breast cancer datasets containing 521 samples that we used as independent validation.mOn the 295 samples, a nearest mean classifier employing a logical OR operation (late integration) on clinical and expression classifiers significantly outperforms all other classifiers. Moreover, regardless of the integration strategy, the nearest mean classifier achieves the best performance. All five classifiers achieve their best performance when integrating clinical and expression data. Repeating the experiments using the 521 samples from the four independent validation datasets also indicated a significant performance improvement when integrating clinical and gene expression data. Whether integration also improves performances on other datasets (e.g. other tumor types) has not been investigated, but seems worthwhile pursuing. Our work suggests that future models for predicting breast cancer outcome should exploit both data types by employing a late OR or intermediate integration strategy based on nearest mean classifiers.

Aurora kinase A outperforms Ki67 as a prognostic marker in ER-positive breast cancer

Background:

Proliferation has emerged as a major prognostic factor in luminal breast cancer. The immunohistochemical (IHC) proliferation marker Ki67 has been most extensively investigated but has not gained widespread clinical acceptance.

Methods:

We have conducted a head-to-head comparison of a panel of proliferation markers, including Ki67. Our aim was to establish the marker of the greatest prognostic utility. Tumour samples from 3093 women with breast cancer were constructed as tissue microarrays. We used IHC to detect expression of mini-chromosome maintenance protein 2, Ki67, aurora kinase A (AURKA), polo-like kinase 1, geminin and phospho-histone H3. We used a Cox proportional-hazards model to investigate the association with 10-year breast cancer-specific survival (BCSS). Missing values were resolved using multiple imputation.

Results:

The prognostic significance of proliferation was limited to oestrogen receptor (ER)-positive breast cancer. Aurora kinase A emerged as the marker of the greatest prognostic significance in a multivariate model adjusted for the standard clinical and molecular covariates (hazard ratio 1.3; 95% confidence interval 1.1–1.5; P=0.005), outperforming all other markers including Ki67.

Conclusion:

Aurora kinase A outperforms other proliferation markers as an independent predictor of BCSS in ER-positive breast cancer. It has the potential for use in routine clinical practice.

A review of the biological and clinical characteristics of luminal-like oestrogen receptor-positive breast cancer

Global gene expression profiling (GEP) studies of breast cancer have identified distinct biological classes with different clinical and therapeutic implications. Oestrogen receptor (ER) has been found to be a central marker of the molecular signature. GEP studies have consistently recognized a molecularly distinct class of tumours that is characterized by high-level expression of ER and other biomarkers recognized to be characteristic of normal luminal cells of the breast. This class is the largest of the GEP-defined molecular subclasses, comprising 60-70% of breast cancer cases. Moreover, it has been proposed that this group of tumours is composed of at least two subclasses distinguished by differing GEP profiles. At present, there is no consensus on the definition of the luminal subclasses and, in clinical practice, luminal-like tumours and ER-positive tumours are frequently considered to be the same. A better understanding of the biological features of luminal tumours could lead to their improved characterization and consistent identification. In this review, we explore the concept and definitions of the luminal-like class of breast carcinoma and their contribution to our understanding of their molecular features, clinical significance and therapeutic implications.

A prognostic model for lymph node-negative breast cancer patients based on the integration of proliferation and immunity

A model for a more precise prognosis of the risk of relapse is needed to avoid overtreatment of lymph node-negative breast cancer patients. A large derivation data set (n = 684) was generated by pooling three independent breast cancer expression microarray data sets. Two major prognostic factors, proliferation and immune response, were identified among genes showing significant differential expression levels between the good outcome and poor outcome groups. For each factor, four proliferation-related genes (p-genes) and four immunity-related genes (i-genes) were selected as prognostic genes, and a prognostic model for lymph node-negative breast cancer patients was developed using a parametric survival analysis based on the lognormal distribution. The p-genes showed a predominantly negative correlation (coefficient: -0.603) with survival time, while the i-genes showed a positive correlation (coefficient: 0.243), reflecting the beneficial effect of the immune response against deleterious proliferative activity. The prognostic model shows that approximately 54% of lymph node-negative breast cancer patients were predicted to be distant metastasis-free for more than 5 years with at least 85% survival probability. The prognostic model showed a robust and high prognostic performance (HR 2.85-3.45) through three external validation data sets. Based on the integration of proliferation and immunity, the new prognostic model is expected to improve clinical decision making by providing easily interpretable survival probabilities at any time point and functional causality of the predicted prognosis with respect to proliferation and immune response.

Integration of pre-normalized microarray data using quantile correction

An enormous amount of microarray data has been collected and accumulated in public repositories. Although some of the depositions include raw and processed data, significant parts of them include processed data only. If we need to combine multiple datasets for specific purposes, the data should be adjusted prior to use to remove bias between the datasets. We focused on a GeneChip platform and a pre-processing method, RMA, and examined simple quantile correction as the post-processing method for integration. Integration of the data pre-processed by RMA was evaluated using artificial spike-in datasets and real microarray datasets of atopic dermatitis and lung cancer. Studies using the spike-in datasets show that the quantile correction for data integration reduces the data quality at some extent but it should be acceptable level. Studies using the real datasets show that the quantile correction significantly reduces the bias. These results show that the quantile correction is useful for integration of multiple datasets processed by RMA, and encourage effective use of public microarray data.

Gene pathways associated with prognosis and chemotherapy sensitivity in molecular subtypes of breast cancer

BACKGROUND

We hypothesized that distinct biological processes might be associated with prognosis and chemotherapy sensitivity in the different types of breast cancers.

METHODS

We performed gene set analyses with BRB-ArrayTools statistical software including 2331 functionally annotated gene sets (ie, lists of genes that correspond to a particular biological pathway or biochemical function) assembled from Ingenuity Pathway Analysis and Gene Ontology databases corresponding to almost all known biological processes. Gene set analysis was performed on gene expression data from three cohorts of 234, 170, and 175 patients with HER2-normal lymph node-negative breast cancer who received no systemic adjuvant therapy to identify gene sets associated prognosis and three additional cohorts of 198, 85, and 62 patients with HER2-normal stage I-III breast cancer who received preoperative chemotherapy to identify gene sets associated with pathological complete response to therapy. These analyses were performed separately for estrogen receptor (ER)-positive and ER-negative breast cancers. Interaction between gene sets and survival and treatment response by breast cancer subtype was assessed in individual datasets and also in pooled datasets. Statistical significance was estimated with permutation test. All statistical tests were two-sided.

RESULTS

For ER-positive cancers, from 370 to 434 gene sets were associated with prognosis (P ≤ .05) and from 209 to 267 gene sets were associated with chemotherapy response in analysis by individual dataset. For ER-positive cancers, 131 gene sets were associated with prognosis and 69 were associated with pathological complete response (P ≤.001) in pooled analysis. Increased expression of cell cycle-related gene sets was associated with poor prognosis, and B-cell immunity-related gene sets were associated with good prognosis. For ER-negative cancers, from 175 to 288 gene sets were associated with prognosis and from 212 to 285 gene sets were associated with chemotherapy response. In pooled analyses of ER-negative cancers, 14 gene sets were associated with prognosis and 23 were associated with response. Gene sets involved in sphingolipid and glycolipid metabolism were associated with better prognosis and those involved in base excision repair, cell aging, and spindle microtubule regulation were associated with chemotherapy response.

CONCLUSION

Different biological processes were associated with prognosis and chemotherapy response in ER-positive and ER-negative breast cancers.

Prognostic gene network modules in breast cancer hold promise

A substantial proportion of lymph node-negative patients who receive adjuvant chemotherapy do not derive any benefit from this aggressive and potentially toxic treatment. However, standard histopathological indices cannot reliably detect patients at low risk of relapse or distant metastasis. In the past few years several prognostic gene expression signatures have been developed and shown to potentially outperform histopathological factors in identifying low-risk patients in specific breast cancer subgroups with predictive values of around 90%, and therefore hold promise for clinical application. We envisage that further improvements and insights may come from integrative expression pathway analyses that dissect prognostic signatures into modules related to cancer hallmarks.

Analysing breast cancer microarrays from African Americans using shrinkage-based discriminant analysis

Breast cancer tumours among African Americans are usually more aggressive than those found in Caucasian populations. African-American patients with breast cancer also have higher mortality rates than Caucasian women. A better understanding of the disease aetiology of these breast cancers can help to improve and develop new methods for cancer prevention, diagnosis and treatment. The main goal of this project was to identify genes that help differentiate between oestrogen receptor-positive and -negative samples among a small group of African-American patients with breast cancer. Breast cancer microarrays from one of the largest genomic consortiums were analysed using 13 African-American and 201 Caucasian samples with oestrogen receptor status. We used a shrinkage-based classification method to identify genes that were informative in discriminating between oestrogen receptor-positive and -negative samples. Subset analysis and permutation were performed to obtain a set of genes unique to the African-American population. We identified a set of 156 probe sets, which gave a misclassification rate of 0.16 in distinguishing between oestrogen receptor-positive and -negative patients. The biological relevance of our findings was explored through literature-mining techniques and pathway mapping. An independent dataset was used to validate our findings and we found that the top ten genes mapped onto this dataset gave a misclassification rate of 0.15. The described method allows us best to utilise the information available from small sample size microarray data in the context of ethnic minorities.

Improved prognostic classification of breast cancer defined by antagonistic activation patterns of immune response pathway modules

Background

Elucidating the activation pattern of molecular pathways across a given tumour type is a key challenge necessary for understanding the heterogeneity in clinical response and for developing novel more effective therapies. Gene expression signatures of molecular pathway activation derived from perturbation experiments in model systems as well as structural models of molecular interactions ("model signatures") constitute an important resource for estimating corresponding activation levels in tumours. However, relatively few strategies for estimating pathway activity from such model signatures exist and only few studies have used activation patterns of pathways to refine molecular classifications of cancer.

Methods

Here we propose a novel network-based method for estimating pathway activation in tumours from model signatures. We find that although the pathway networks inferred from cancer expression data are highly consistent with the prior information contained in the model signatures, that they also exhibit a highly modular structure and that estimation of pathway activity is dependent on this modular structure. We apply our methodology to a panel of 438 estrogen receptor negative (ER-) and 785 estrogen receptor positive (ER+) breast cancers to infer activation patterns of important cancer related molecular pathways.

Results

We show that in ER negative basal and HER2+ breast cancer, gene expression modules reflecting T-cell helper-1 (Th1) and T-cell helper-2 (Th2) mediated immune responses play antagonistic roles as major risk factors for distant metastasis. Using Boolean interaction Cox-regression models to identify non-linear pathway combinations associated with clinical outcome, we show that simultaneous high activation of Th1 and low activation of a TGF-beta pathway module defines a subtype of particularly good prognosis and that this classification provides a better prognostic model than those based on the individual pathways. In ER+ breast cancer, we find that simultaneous high MYC and RAS activity confers significantly worse prognosis than either high MYC or high RAS activity alone. We further validate these novel prognostic classifications in independent sets of 173 ER- and 567 ER+ breast cancers.

Conclusion

We have proposed a novel method for pathway activity estimation in tumours and have shown that pathway modules antagonize or synergize to delineate novel prognostic subtypes. Specifically, our results suggest that simultaneous modulation of T-helper differentiation and TGF-beta pathways may improve clinical outcome of hormone insensitive breast cancers over treatments that target only one of these pathways.

Nuclear receptor coregulators as a new paradigm for therapeutic targeting

The complex function and regulation of nuclear receptors cannot be fully understood without a thorough knowledge of the receptor-associated coregulators that either enhance (coactivators) or inhibit (corepressors) transcription. While nuclear receptors themselves have garnered much attention as therapeutic targets, the clinical and etiological relevance of the coregulators to human diseases is increasingly recognized. Aberrant expression or function of coactivators and corepressors has been associated with malignant and metabolic disease development. Many of them are key epigenetic regulators and utilize enzymatic activities to modify chromatin through histone acetylation/deacetylation, histone methylation/demethylation or chromatin remodeling. In this review, we showcase and evaluate coregulators--such as SRCs and ANCCA--with the most promising therapeutic potential based on their physiological roles and involvement in various diseases that are revealed thus far. We also describe the structural features of the coactivator and corepressor functional domains and highlight areas that can be further explored for molecular targeting.

ANCCA/ATAD2 overexpression identifies breast cancer patients with poor prognosis, acting to drive proliferation and survival of triple-negative cells through control of B-Myb and EZH2

Chromatin coregulators are important factors in tumorigenesis and cancer progression. ANCCA is an AAA+ ATPase and a bromodomain-containing nuclear coactivator for the estrogen and androgen receptors that is crucial for assembly of chromatin-modifying complexes and proliferation of hormone-responsive cancer cells. In this study, we show that ANCCA is overexpressed in >70% of breast tumors and that its high protein level correlates well with tumor histologic grades (P<0.0001), highlighting ANCCA as a prognostic factor for poor overall survival and disease recurrence. Strikingly, high-level ANCCA correlated with triple-negative tumors that represent highly aggressive disease. Analysis of ANCCA transcript levels in multiple expression profiles of breast cancer identified ANCCA as a common signature gene, indicating that elevated transcripts also strongly correlate with tumor metastasis and poor survival. Biological and mechanistic investigations revealed that ANCCA is crucial for proliferation and survival of triple-negative/basal-like cancer cells and that it controls the expression of B-Myb, histone methyltransferase EZH2, and an Rb-E2F core program for proliferation, along with a subset of key mitotic kinesins and cell survival genes (IRS2, VEGF, and Akt1). In particular, ANCCA overexpression correlated strongly with EZH2 in tumors. Our results suggest that ANCCA may integrate multiple oncogenic programs in breast cancer, serving in particular as a prognostic marker and a therapeutic target for triple-negative cancers.

Gene expression profile predicts outcome after anthracycline-based adjuvant chemotherapy in early breast cancer

Prognosis of early beast cancer is heterogeneous. Today, no histoclinical or biological factor predictive for clinical outcome after adjuvant anthracycline-based chemotherapy (CT) has been validated and introduced in routine use. Using DNA microarrays, we searched for a gene expression signature associated with metastatic relapse after adjuvant anthracycline-based CT without taxane. We profiled a multicentric series of 595 breast cancers including 498 treated with such adjuvant CT. The identification of the prognostic signature was done using a metagene-based supervised approach in a learning set of 323 patients. The signature was then tested on an independent validation set comprising 175 similarly treated patients, 128 of them from the PACS01 prospective clinical trial. We identified a 3-metagene predictor of metastatic relapse in the learning set, and confirmed its independent prognostic impact in the validation set. In multivariate analysis, the predictor outperformed the individual current prognostic factors, as well as the Nottingham Prognostic Index-based classifier, both in the learning and the validation sets, and added independent prognostic information. Among the patients treated with adjuvant anthracycline-based CT, with a median follow-up of 68 months, the 5-year metastasis-free survival was 82% in the "good-prognosis" group and 56% in the "poor-prognosis" group. Our predictor refines the prediction of metastasis-free survival after adjuvant anthracycline-based CT and might help tailoring adjuvant CT regimens.

Prediction of breast cancer prognosis using gene set statistics provides signature stability and biological context

Background

Different microarray studies have compiled gene lists for predicting outcomes of a range of treatments and diseases. These have produced gene lists that have little overlap, indicating that the results from any one study are unstable. It has been suggested that the underlying pathways are essentially identical, and that the expression of gene sets, rather than that of individual genes, may be more informative with respect to prognosis and understanding of the underlying biological process.

Results

We sought to examine the stability of prognostic signatures based on gene sets rather than individual genes. We classified breast cancer cases from five microarray studies according to the risk of metastasis, using features derived from predefined gene sets. The expression levels of genes in the sets are aggregated, using what we call a set statistic. The resulting prognostic gene sets were as predictive as the lists of individual genes, but displayed more consistent rankings via bootstrap replications within datasets, produced more stable classifiers across different datasets, and are potentially more interpretable in the biological context since they examine gene expression in the context of their neighbouring genes in the pathway. In addition, we performed this analysis in each breast cancer molecular subtype, based on ER/HER2 status. The prognostic gene sets found in each subtype were consistent with the biology based on previous analysis of individual genes.

Conclusions

To date, most analyses of gene expression data have focused at the level of the individual genes. We show that a complementary approach of examining the data using predefined gene sets can reduce the noise and could provide increased insight into the underlying biological pathways.

Gene expression profiling of response to mTOR inhibitor everolimus in pre-operatively treated post-menopausal women with oestrogen receptor-positive breast cancer

There is growing evidence that uncontrolled activation of the PI3K/Akt/mTOR pathway contributes to the development and progression of breast cancer. Inhibition of this pathway has antitumour effects in preclinical studies and efficacy in combination with other agents in breast cancer patients. The aim of this study is to characterise the effects of pre-operative everolimus treatment in primary breast cancer patients and to identify potential molecular predictors of response. Twenty-seven patients with oestrogen receptor (ER)-positive breast cancer completed 11-14 days of neoadjuvant treatment with 5-mg everolimus. Core biopsies were taken before and after treatment and analysed using Illumina HumanRef-8 v2 Expression BeadChips. Changes in proliferation (Ki67) and phospho-AKT were measured on diagnostic core biopsies/resection samples embedded in paraffin by immunohistochemistry to determine response to treatment. Patients that responded to everolimus treatment with significant reductions in proliferation (fall in % Ki67 positive cells) also had significant decreases in the expression of genes involved in cell cycle (P = 8.70E-09) and p53 signalling (P = 0.01) pathways. Highly proliferating tumours that have a poor prognosis exhibited dramatic reductions in the expression of cell cycle genes following everolimus treatment. The genes that most clearly separated responding from non-responding pre-treatment tumours were those involved with protein modification and dephosphorylation, including DYNLRB2, ERBB4, PTPN13, ULK2 and DUSP16. The majority of ER-positive breast tumours treated with everolimus showed a significant reduction in genes involved with proliferation, these may serve as markers of response and predict which patients will derive most benefit from mTOR inhibition.

Prediction of adjuvant chemotherapy benefit in endocrine responsive, early breast cancer using multigene assays

BACKGROUND

Multigene assays performed on the primary tumors from women with non-metastatic breast cancer provide useful prognostic information and discriminate excellent versus poor outcome potential in diverse clinical scenarios. Recently, analyses were conducted to determine if these assays predict who benefits from adjuvant chemotherapy added to endocrine therapy and conversely, who might avoid chemotherapy because of lack of substantial benefit. This literature-based review summarizes these data and provides a perspective on the limitations and clinical utility of these assays.

METHODS

The literature regarding multigene assays and signatures in early breast cancer was surveyed. Only two assays-- the 21-gene recurrence score (RS) assay (Oncotype DX) and the 70-gene signature (MammaPrint)--were analyzed in randomized or non-randomized clinical populations in order to determine the predictive utility of the test in the adjuvant chemotherapy setting in patients whose tumors were estrogen-receptor positive. These data are summarized by type of clinical analysis, with information on clinical utility and comparative studies with standard clinical-pathologic factors.

RESULTS

From 2 independent analyses in phase III clinical trial settings with tamoxifen-alone control arms, the 21-gene RS assay defines a group of patients with low scores who do not appear to benefit from chemotherapy, and a second group with very high scores who derive major benefit from CMF or CAF chemotherapy. One study was conducted in node-negative disease, and the second in a node-positive population. Interaction terms were significant in both studies, and the effect of the assay remained upon adjustment for other standard factors. Utilizing a non-randomized clinical setting, the 70-gene signature could also predict chemotherapy benefit in the high risk group, versus no apparent benefit in the low risk group, an effect that remained after adjustment for standard factors. For both assays, the discordance rate between the assay prediction and clinical-pathologic risk category was approximately 30%. Clinical utility studies showed use of the assay results in a change in treatment decision in 25-30% of cases, most commonly from chemoendocrine therapy to endocrine therapy alone.

SUMMARY

The prediction of adjuvant chemotherapy benefit over and above endocrine therapy using multigene assay-determined risk category differs greatly across risk level and challenges the previous adjuvant therapy paradigm that degree of benefit is the same regardless of risk. These data justify current clinical use of these assays, while ongoing prospective studies will refine their role in practice settings.

Correcting for intra-experiment variation in Illumina BeadChip data is necessary to generate robust gene-expression profiles

Background

Microarray technology is a popular means of producing whole genome transcriptional profiles, however high cost and scarcity of mRNA has led many studies to be conducted based on the analysis of single samples. We exploit the design of the Illumina platform, specifically multiple arrays on each chip, to evaluate intra-experiment technical variation using repeated hybridisations of universal human reference RNA (UHRR) and duplicate hybridisations of primary breast tumour samples from a clinical study.

Results

A clear batch-specific bias was detected in the measured expressions of both the UHRR and clinical samples. This bias was found to persist following standard microarray normalisation techniques. However, when mean-centering or empirical Bayes batch-correction methods (ComBat) were applied to the data, inter-batch variation in the UHRR and clinical samples were greatly reduced. Correlation between replicate UHRR samples improved by two orders of magnitude following batch-correction using ComBat (ranging from 0.9833-0.9991 to 0.9997-0.9999) and increased the consistency of the gene-lists from the duplicate clinical samples, from 11.6% in quantile normalised data to 66.4% in batch-corrected data. The use of UHRR as an inter-batch calibrator provided a small additional benefit when used in conjunction with ComBat, further increasing the agreement between the two gene-lists, up to 74.1%.

Conclusion

In the interests of practicalities and cost, these results suggest that single samples can generate reliable data, but only after careful compensation for technical bias in the experiment. We recommend that investigators appreciate the propensity for such variation in the design stages of a microarray experiment and that the use of suitable correction methods become routine during the statistical analysis of the data.

Survival prediction from clinico-genomic models--a comparative study

Background

Survival prediction from high-dimensional genomic data is an active field in today's medical research. Most of the proposed prediction methods make use of genomic data alone without considering established clinical covariates that often are available and known to have predictive value. Recent studies suggest that combining clinical and genomic information may improve predictions, but there is a lack of systematic studies on the topic. Also, for the widely used Cox regression model, it is not obvious how to handle such combined models.

Results

We propose a way to combine classical clinical covariates with genomic data in a clinico-genomic prediction model based on the Cox regression model. The prediction model is obtained by a simultaneous use of both types of covariates, but applying dimension reduction only to the high-dimensional genomic variables. We describe how this can be done for seven well-known prediction methods: variable selection, unsupervised and supervised principal components regression and partial least squares regression, ridge regression, and the lasso. We further perform a systematic comparison of the performance of prediction models using clinical covariates only, genomic data only, or a combination of the two. The comparison is done using three survival data sets containing both clinical information and microarray gene expression data. Matlab code for the clinico-genomic prediction methods is available at http://www.med.uio.no/imb/stat/bmms/software/clinico-genomic/.

Conclusions

Based on our three data sets, the comparison shows that established clinical covariates will often lead to better predictions than what can be obtained from genomic data alone. In the cases where the genomic models are better than the clinical, ridge regression is used for dimension reduction. We also find that the clinico-genomic models tend to outperform the models based on only genomic data. Further, clinico-genomic models and the use of ridge regression gives for all three data sets better predictions than models based on the clinical covariates alone.

Common germline polymorphisms in COMT, CYP19A1, ESR1, PGR, SULT1E1 and STS and survival after a diagnosis of breast cancer

Although preliminary evidence suggests that germline variation in genes involved in steroid hormone metabolism may alter breast cancer prognosis, this has not been systematically evaluated. We examined associations between germline polymorphisms in 6 genes involved in the steroid hormone metabolism and signaling pathway (COMT, CYP19A1, ESR1, PGR, SULT1E1, STS) and survival among women with breast cancer participating in SEARCH, a population-based case-control study. Blood samples from up to 4,470 women were genotyped for 4 possible functional SNPs in CYP19A1 and 106 SNPs tagging the common variation in the remainder of the genes. The genotypes of each polymorphism were tested for association with survival after breast cancer diagnosis using Cox regression analysis. Significant evidence of an association was observed for a COMT polymorphism (rs4818 p = 0.016) under the codominant model. This SNP appeared to fit a dominant model better (HR = 0.80 95% CI: 0.69-0.95, p = 0.009); however, the result was only marginally significant after permutation analysis adjustment for multiple hypothesis tests (p = 0.047). To further evaluate this finding, somatic expression microarray data from 8 publicly available datasets were used to test the association between survival and tumor COMT gene expression; no statistically significant associations were observed. A correlated SNP in COMT, rs4860, has recently been associated with breast cancer prognosis in Chinese women in a dominant model. These results suggest that COMT rs4818, or a variant it tags, is associated with breast cancer prognosis. Further study of COMT and its putative association with breast cancer prognosis is warranted.

Immutable functional attributes of histologic grade revealed by context-independent gene expression in primary breast cancer cells

Inherent cancer phenotypes that are independent of fluctuating cross-talk with the surrounding tissue matrix are highly desirable candidates for targeting tumor cells. Our novel study design uses epithelial cell lines derived from low versus high histologic grade primary breast cancer to effectively diminish the breadth of transient variability generated within the tumor microenvironment of the host, revealing a "paracrine-independent expression of grade-associated" (PEGA) gene signature. PEGA members extended beyond "proliferation-driven" signatures commonly associated with aggressive, high-grade breast cancer. The calcium-binding protein S100P was prominent among PEGA genes overexpressed in high-grade tumors. A three-member fingerprint of S100P-correlated genes, consisting of GPRC5A, FXYD3, and PYCARD, conferred poor outcome in multiple breast cancer data sets, irrespective of estrogen receptor status but dependent on tumor size (P < 0.01). S100P silencing markedly diminished coregulated gene transcripts and reversed aggressive tumor behavior. Exposure to pathway-implicated agents, including the calmodulin inhibitor N-(6-aminohexyl)-5-chloro-1-naphthalenesulfonamide, phenothiazine, and chlorpromazine, resulted in rapid apoptotic cell death in high-grade tumor cells resistant to the chemotherapeutic drug cisplatin. This is the first comprehensive study describing molecular phenotypes intimately associated with histologic grade whose expression remains relatively fixed despite an unavoidably changing environment to which tumor cells are invariably exposed.

A systems biology-based gene expression classifier of glioblastoma predicts survival with solid tumors

Accurate prediction of survival of cancer patients is still a key open problem in clinical research. Recently, many large-scale gene expression clusterings have identified sets of genes reportedly predictive of prognosis; however, those gene sets shared few genes in common and were poorly validated using independent data. We have developed a systems biology-based approach by using either combined gene sets and the protein interaction network (Method A) or the protein network alone (Method B) to identify common prognostic genes based on microarray gene expression data of glioblastoma multiforme and compared with differential gene expression clustering (Method C). Validations of prediction performance show that the 23-prognostic gene classifier identified by Method A outperforms other gene classifiers identified by Methods B and C or previously reported for gliomas on 17 of 20 independent sample cohorts across five tumor types. We also find that among the 23 genes are 21 related to cellular proliferation and two related to response to stress/immune response. We further find that the increased expression of the 21 genes and the decreased expression of the other two genes are associated with poorer survival, which is supportive with the notion that cellular proliferation and immune response contribute to a significant portion of predictive power of prognostic classifiers. Our results demonstrate that the systems biology-based approach enables to identify common survival-associated genes.

Intrinsic bias in breast cancer gene expression data sets

Background

While global breast cancer gene expression data sets have considerable commonality in terms of their data content, the populations that they represent and the data collection methods utilized can be quite disparate. We sought to assess the extent and consequence of these systematic differences with respect to identifying clinically significant prognostic groups.

Methods

We ascertained how effectively unsupervised clustering employing randomly generated sets of genes could segregate tumors into prognostic groups using four well-characterized breast cancer data sets.

Results

Using a common set of 5,000 randomly generated lists (70 genes/list), the percentages of clusters with significant differences in metastasis latencies (HR p-value < 0.01) was 62%, 15%, 21% and 0% in the NKI2 (Netherlands Cancer Institute), Wang, TRANSBIG and KJX64/KJ125 data sets, respectively. Among ER positive tumors, the percentages were 38%, 11%, 4% and 0%, respectively. Few random lists were predictive among ER negative tumors in any data set. Clustering was associated with ER status and, after globally adjusting for the effects of ER-α gene expression, the percentages were 25%, 33%, 1% and 0%, respectively. The impact of adjusting for ER status depended on the extent of confounding between ER-α gene expression and markers of proliferation.

Conclusion

It is highly probable to identify a statistically significant association between a given gene list and prognosis in the NKI2 dataset due to its large sample size and the interrelationship between ER-α expression and markers of proliferation. In most respects, the TRANSBIG data set generated similar outcomes as the NKI2 data set, although its smaller sample size led to fewer statistically significant results.

Effects of sample size on robustness and prediction accuracy of a prognostic gene signature

BACKGROUND

Few overlap between independently developed gene signatures and poor inter-study applicability of gene signatures are two of major concerns raised in the development of microarray-based prognostic gene signatures. One recent study suggested that thousands of samples are needed to generate a robust prognostic gene signature.

RESULTS

A data set of 1,372 samples was generated by combining eight breast cancer gene expression data sets produced using the same microarray platform and, using the data set, effects of varying samples sizes on a few performances of a prognostic gene signature were investigated. The overlap between independently developed gene signatures was increased linearly with more samples, attaining an average overlap of 16.56% with 600 samples. The concordance between predicted outcomes by different gene signatures also was increased with more samples up to 94.61% with 300 samples. The accuracy of outcome prediction also increased with more samples. Finally, analysis using only Estrogen Receptor-positive (ER+) patients attained higher prediction accuracy than using both patients, suggesting that sub-type specific analysis can lead to the development of better prognostic gene signatures

CONCLUSION

Increasing sample sizes generated a gene signature with better stability, better concordance in outcome prediction, and better prediction accuracy. However, the degree of performance improvement by the increased sample size was different between the degree of overlap and the degree of concordance in outcome prediction, suggesting that the sample size required for a study should be determined according to the specific aims of the study.

Bioinformatics and breast cancer: what can high-throughput genomic approaches actually tell us?

High-throughput genomic technology has rapidly become a major tool for the study of breast cancer. Gene expression profiling has been applied to many areas of research from basic science to translational studies, with the potential to identify new targets for treatment, mechanisms of resistance and to improve on current tools for the analysis of prognosis. However, the sheer scale of the data generated along with the number of different protocols, platforms and analysis methods can make these studies difficult for clinicians to comprehend. Similarly, computational scientists and statisticians that may be called upon to analyse the data generated are often unaware of the processes involved in sample collection or the relevance and impact of genetics and pathological characteristics. There is a pressing need for better understanding of the challenges and limitations of microarray approaches, both in experimental design and data analysis. Holistic, whole-genome approaches are still relatively new and critics have been quick to highlight non-overlapping results from groups testing similar hypotheses. However, it is often subtle differences in the experimental design and technology that underpin the variation between these studies. Rather than indicating that the data are meaningless, this suggests that many findings are real, but highly context dependent. This review explores both the current state and potential of bioinformatics to bring meaning to high-throughput genomic approaches in the understanding of breast cancer.

Approaches towards expression profiling the response to treatment

Over the past 8 years there has been a wealth of breast cancer gene expression studies. The majority of these studies have focused upon characterising a tumour at presentation, before treatment, rather than looking at the effects of treatment on the tumour. More recently, a number of groups have moved from predicting prognosis based upon long-term follow-up to alternative approaches of using expression profiling to measure the effect of treatment on breast tumours and potentially predict response to therapy using either post-treatment samples or both pre-treatment and post-treatment samples. Whilst this provides great potential to further our understanding of the mode of action of treatments and to more accurately select which patients will benefit from a particular treatment, serious issues of experimental design must be considered.