The effect of the stromal component of breast tumours on prediction of clinical outcome using gene expression microarray analysis

Automated Prognosis Marker Assessment in Breast Cancers Using BLEACH&STAIN Multiplexed Immunohistochemistry

Prognostic markers in routine clinical management of breast cancer are often assessed using RNA-based multi-gene panels that depend on fluctuating tumor purity. Multiplex fluorescence immunohistochemistry (mfIHC) holds the potential for an improved risk assessment. To enable automated prognosis marker detection (i.e., progesterone receptor [PR], estrogen receptor [ER], androgen receptor [AR], GATA3, TROP2, HER2, PD-L1, Ki67, TOP2A), a framework for automated breast cancer identification was developed and validated involving thirteen different artificial intelligence analysis steps and an algorithm for cell distance analysis using 11+1-marker-BLEACH&STAIN-mfIHC staining in 1404 invasive breast cancers of no special type (NST). The framework for automated breast cancer detection discriminated normal glands from malignant glands with an accuracy of 98.4%. This approach identified that five (PR, ER, AR, GATA3, PD-L1) of nine biomarkers were associated with prolonged overall survival (p ≤ 0.0095 each) and two of these (PR, AR) were found to be independent risk factors in multivariate analysis (p ≤ 0.0151 each). The combined assessment of PR-ER-AR-GATA3-PD-L1 as a five-marker prognosis score showed strong prognostic relevance (p < 0.0001) and was an independent risk factor in multivariate analysis (p = 0.0034). Automated breast cancer detection in combination with an artificial intelligence-based analysis of mfIHC enables a rapid and reliable analysis of multiple prognostic parameters. The strict limitation of the analysis to malignant cells excludes the impact of fluctuating tumor purity on assay precision.

Systematic in vitro analysis of therapy resistance in glioblastoma cell lines by integration of clonogenic survival data with multi-level molecular data

Despite intensive basic scientific, translational, and clinical efforts in the last decades, glioblastoma remains a devastating disease with a highly dismal prognosis. Apart from the implementation of temozolomide into the clinical routine, novel treatment approaches have largely failed, emphasizing the need for systematic examination of glioblastoma therapy resistance in order to identify major drivers and thus, potential vulnerabilities for therapeutic intervention. Recently, we provided proof-of-concept for the systematic identification of combined modality radiochemotherapy treatment vulnerabilities via integration of clonogenic survival data upon radio(chemo)therapy with low-density transcriptomic profiling data in a panel of established human glioblastoma cell lines. Here, we expand this approach to multiple molecular levels, including genomic copy number, spectral karyotyping, DNA methylation, and transcriptome data. Correlation of transcriptome data with inherent therapy resistance on the single gene level yielded several candidates that were so far underappreciated in this context and for which clinically approved drugs are readily available, such as the androgen receptor (AR). Gene set enrichment analyses confirmed these results, and identified additional gene sets, including reactive oxygen species detoxification, mammalian target of rapamycin complex 1 (MTORC1) signaling, and ferroptosis/autophagy-related regulatory circuits to be associated with inherent therapy resistance in glioblastoma cells. To identify pharmacologically accessible genes within those gene sets, leading edge analyses were performed yielding candidates with functions in thioredoxin/peroxiredoxin metabolism, glutathione synthesis, chaperoning of proteins, prolyl hydroxylation, proteasome function, and DNA synthesis/repair. Our study thus confirms previously nominated targets for mechanism-based multi-modal glioblastoma therapy, provides proof-of-concept for this workflow of multi-level data integration, and identifies novel candidates for which pharmacological inhibitors are readily available and whose targeting in combination with radio(chemo)therapy deserves further examination. In addition, our study also reveals that the presented workflow requires mRNA expression data, rather than genomic copy number or DNA methylation data, since no stringent correlation between these data levels could be observed. Finally, the data sets generated in the present study, including functional and multi-level molecular data of commonly used glioblastoma cell lines, represent a valuable toolbox for other researchers in the field of glioblastoma therapy resistance.

Spatially-resolved quantification of proteins in triple negative breast cancers reveals differences in the immune microenvironment associated with prognosis

Triple negative breast cancer (TNBC) is an aggressive breast cancer subtype. Recent studies have shown that MHC class II (MHCII) expression and tumor infiltrating lymphocytes are important prognostic factors in patients with TNBC, although the relative importance of lymphocyte subsets and associated protein expression is incompletely understood. NanoString Digital Spatial Profiling (DSP) allows for spatially resolved, highly multiplexed quantification of proteins in clinical samples. In this study, we sought to determine if DSP could be used to characterize expression of MHCII and other immune related proteins in tumor epithelial versus stromal compartments of patient-derived TNBCs (N = 10) using a panel of 39 markers. We confirmed that a subset of TNBCs have elevated expression of HLA-DR in tumor epithelial cells; HLA-DR expression was also significantly higher in the tumors of patients with long-term disease-free survival when compared to patients that relapsed. HLA-DR expression in the epithelial compartment was correlated with high expression of CD4 and ICOS in the stromal compartment of the same tumors. We also identified candidate protein biomarkers with significant differential expression between patients that relapsed versus those that did not. In conclusion, DSP is a powerful method that allows for quantification of proteins in the immune microenvironment of TNBCs.

Schlafen-11 expression is associated with immune signatures and basal-like phenotype in breast cancer

PURPOSE

Breast cancer (BC) is a heterogeneous disorder, with variable response to systemic chemotherapy. Likewise, BC shows highly complex immune activation patterns, only in part reflecting classical histopathological subtyping. Schlafen-11 (SLFN11) is a nuclear protein we independently described as causal factor of sensitivity to DNA damaging agents (DDA) in cancer cell line models. SLFN11 has been reported as a predictive biomarker for DDA and PARP inhibitors in human neoplasms. SLFN11 has been implicated in several immune processes such as thymocyte maturation and antiviral response through the activation of interferon signaling pathway, suggesting its potential relevance as a link between immunity and cancer. In the present work, we investigated the transcriptional landscape of SLFN11, its potential prognostic value, and the clinico-pathological associations with its variability in BC.

METHODS

We assessed SLFN11 determinants in a gene expression meta-set of 5061 breast cancer patients annotated with clinical data and multigene signatures.

RESULTS

We found that 537 transcripts are highly correlated with SLFN11, identifying "immune response", "lymphocyte activation", and "T cell activation" as top Gene Ontology processes. We established a strong association of SLFN11 with stromal signatures of basal-like phenotype and response to chemotherapy in estrogen receptor negative (ER-) BC. We identified a distinct subgroup of patients, characterized by high SLFN11 levels, ER- status, basal-like phenotype, immune activation, and younger age. Finally, we observed an independent positive predictive role for SLFN11 in BC.

CONCLUSIONS

Our findings are suggestive of a relevant role for SLFN11 in BC and its immune and molecular variability.

A sequential Monte Carlo approach to gene expression deconvolution

High-throughput gene expression data are often obtained from pure or complex (heterogeneous) biological samples. In the latter case, data obtained are a mixture of different cell types and the heterogeneity imposes some difficulties in the analysis of such data. In order to make conclusions on gene expresssion data obtained from heterogeneous samples, methods such as microdissection and flow cytometry have been employed to physically separate the constituting cell types. However, these manual approaches are time consuming when measuring the responses of multiple cell types simultaneously. In addition, exposed samples, on many occasions, end up being contaminated with external perturbations and this may result in an altered yield of molecular content. In this paper, we model the heterogeneous gene expression data using a Bayesian framework, treating the cell type proportions and the cell-type specific expressions as the parameters of the model. Specifically, we present a novel sequential Monte Carlo (SMC) sampler for estimating the model parameters by approximating their posterior distributions with a set of weighted samples. The SMC framework is a robust and efficient approach where we construct a sequence of artificial target (posterior) distributions on spaces of increasing dimensions which admit the distributions of interest as marginals. The proposed algorithm is evaluated on simulated datasets and publicly available real datasets, including Affymetrix oligonucleotide arrays and national center for biotechnology information (NCBI) gene expression omnibus (GEO), with varying number of cell types. The results obtained on all datasets show a superior performance with an improved accuracy in the estimation of cell type proportions and the cell-type specific expressions, and in addition, more accurate identification of differentially expressed genes when compared to other widely known methods for blind decomposition of heterogeneous gene expression data such as Dsection and the nonnegative matrix factorization (NMF) algorithms. MATLAB implementation of the proposed SMC algorithm is available to download at https://github.com/moyanre/smcgenedeconv.git.

Standardising RNA profiling based biomarker application in cancer-The need for robust control of technical variables

Histopathology-based staging of colorectal cancer (CRC) has utility in assessing the prognosis of patient subtypes, but as yet cannot accurately predict individual patient's treatment response. Transcriptomics approaches, using array based or next generation sequencing (NGS) platforms, of formalin fixed paraffin embedded tissue can be harnessed to develop multi-gene biomarkers for predicting both prognosis and treatment response, leading to stratification of treatment. While transcriptomics can shape future biomarker development, currently <1% of published biomarkers become clinically validated tests, often due to poor study design or lack of independent validation. In this review of a large number of CRC transcriptional studies, we identify recurrent sources of technical variability that encompass collection, preservation and storage of malignant tissue, nucleic acid extraction, methods to quantitate RNA transcripts and data analysis pipelines. We propose a series of defined steps for removal of these confounding issues, to ultimately aid in the development of more robust clinical biomarkers.

Accurate prediction of response to endocrine therapy in breast cancer patients: current and future biomarkers

Approximately 70% of patients have breast cancers that are oestrogen receptor alpha positive (ER+) and are therefore candidates for endocrine treatment. Many of these patients relapse in the years during or following completion of adjuvant endocrine therapy. Thus, many ER+ cancers have primary resistance or develop resistance to endocrine therapy during treatment. Recent improvements in our understanding of how tumours evolve during treatment with endocrine agents have identified both changes in gene expression and mutational profiles, in the primary cancer as well as in circulating tumour cells. Analysing these changes has the potential to improve the prediction of which specific patients will respond to endocrine treatment. Serially profiled biopsies during treatment in the neoadjuvant setting offer promise for accurate and early prediction of response to both current and novel drugs and allow investigation of mechanisms of resistance. In addition, recent advances in monitoring tumour evolution through non-invasive (liquid) sampling of circulating tumour cells and cell-free tumour DNA may provide a method to detect resistant clones and allow implementation of personalized treatments for metastatic breast cancer patients. This review summarises current and future biomarkers and signatures for predicting response to endocrine treatment, and discusses the potential for using approved drugs and novel agents to improve outcomes. Increased prediction accuracy is likely to require sequential sampling, utilising preoperative or neoadjuvant treatment and/or liquid biopsies and an improved understanding of both the dynamics and heterogeneity of breast cancer.

Analysis of Altered Micro RNA Expression Profiles in Focal Cortical Dysplasia IIB

Focal cortical dysplasia type IIB is a commonly encountered subtype of developmental malformation of the cerebral cortex and is often associated with pharmacoresistant epilepsy. In this study, to investigate the molecular etiology of focal cortical dysplasia type IIB, the authors performed micro ribonucleic acid (RNA) microarray on surgical specimens from 5 children (2 female and 3 male, mean age was 73.4 months, range 50-112 months) diagnosed of focal cortical dysplasia type IIB and matched normal tissue adjacent to the lesion. In all, 24 micro RNAs were differentially expressed in focal cortical dysplasia type IIB, and the microarray results were validated using quantitative real-time polymerase chain reaction (PCR). Then the putative target genes of the differentially expressed micro RNAs were identified by bioinformatics analysis. Moreover, biological significance of the target genes was evaluated by investigating the pathways in which the genes were enriched, and the Hippo signaling pathway was proposed to be highly related with the pathogenesis of focal cortical dysplasia type IIB.

Comparison of targeted next-generation sequencing and Sanger sequencing for the detection of PIK3CA mutations in breast cancer

Background

Phosphatidylinositol-4,5-bisphosphate 3-kinase, catalytic subunit alpha, PIK3CA, is one of the most frequently mutated genes in breast cancer, and the mutation status of PIK3CA has clinical relevance related to response to therapy.

The aim of our study was to investigate the mutation status of PIK3CA gene and to evaluate the concordance between NGS and SGS for the most important hotspot regions in exon 9 and 20, to investigate additional hotspots outside of these exons using NGS, and to correlate the PIK3CA mutation status with the clinicopathological characteristics of the cohort.

Methods

In the current study, next-generation sequencing (NGS) and Sanger Sequencing (SGS) was used for the mutational analysis of PIK3CA in 186 breast carcinomas.

Results

Altogether, 64 tumors had PIK3CA mutations, 55 of these mutations occurred in exons 9 and 20. Out of these 55 mutations, 52 could also be detected by Sanger sequencing resulting in a concordance of 98.4 % between the two sequencing methods. The three mutations missed by SGS had low variant frequencies below 10 %. Additionally, 4.8 % of the tumors had mutations in exons 1, 4, 7, and 13 of PIK3CA that were not detected by SGS. PIK3CA mutation status was significantly associated with hormone receptor-positivity, HER2-negativity, tumor grade, and lymph node involvement. However, there was no statistically significant association between the PIK3CA mutation status and overall survival.

Conclusions

Based on our study, NGS is recommended as follows: 1) for correctly assessing the mutation status of PIK3CA in breast cancer, especially for cases with low tumor content, 2) for the detection of subclonal mutations, and 3) for simultaneous mutation detection in multiple exons.

Electronic supplementary material

The online version of this article (doi:10.1186/s12907-015-0020-6) contains supplementary material, which is available to authorized users.

Prognostic stromal gene signatures in breast cancer

Introduction

Global gene expression analysis of tumor samples has been a valuable tool to subgroup tumors and has the potential to be of prognostic and predictive value. However, tumors are heterogeneous, and homogenates will consist of several different cell types. This study was designed to obtain more refined expression data representing different compartments of the tumor.

Methods

Formalin-fixed paraffin-embedded stroma-rich triple-negative breast cancer tumors were laser-microdissected, and RNA was extracted and processed to enable microarray hybridization. Genes enriched in stroma were identified and used to generate signatures by identifying correlating genes in publicly available data sets. The prognostic implications of the signature were analyzed.

Results

Comparison of the expression pattern from stromal and cancer cell compartments from three tumors revealed a number of genes that were essentially specifically expressed in the respective compartments. The stroma-specific genes indicated contribution from fibroblasts, endothelial cells, and immune/inflammatory cells. The gene set was expanded by identifying correlating mRNAs using breast cancer mRNA expression data from The Cancer Genome Atlas. By iterative analyses, 16 gene signatures of highly correlating genes were characterized. Based on the gene composition, they seem to represent different cell types. In multivariate Cox proportional hazard models, two immune/inflammatory signatures had opposing hazard ratios for breast cancer recurrence also after adjusting for clinicopathological variables and molecular subgroup. The signature associated with poor prognosis consisted mainly of C1Q genes and the one associated with good prognosis contained HLA genes. This association with prognosis was seen for other cancers as well as in other breast cancer data sets.

Conclusions

Our data indicate that the molecular composition of the immune response in a tumor may be a powerful predictor of cancer prognosis.

Electronic supplementary material

The online version of this article (doi:10.1186/s13058-015-0530-2) contains supplementary material, which is available to authorized users.

Profile of differentially expressed intratumoral cytokines to predict the immune-polarizing side effects of tamoxifen in breast cancer treatment

Factors within the tissue of breast cancer (BC) may shift the polarization of CD4+ T cells towards Th2 direction. This tendency can promote tumor development and be enhanced by the use of tamoxifen during the treatment. Thus, the patients with low levels of tumor-induced Th2 polarization prior to tamoxifen treatment may better endure the immune-polarizing side effects (IPSE) of tamoxifen and have better prognoses. Estimation of Th2 polarization status should help predict the IPSE among tamoxifen-treated patients and guide the use of tamoxifen among all BC patients before the tamoxifen therapy. Here, we report profiling of differentially expressed (DE) intratumoral cytokines as a signature to evaluate the IPSE of tamoxifen. The DE genes of intratumoral CD4+ T cells (CD4 DEGs) were identified by gene expression profiles of purified CD4+ T cells from BC patients and validated by profiling of cultured intratumoral CD4+ T cells. Functional enrichment analyses showed a directed Th2 polarization of intratumoral CD4+ T cells. To find the factors inducing the Th2 polarization of CD4+ T cells, we identified 995 common DE genes of bulk BC tissues (BC DEGs) by integrating five independent datasets. Five DE cytokines observed in bulk BC tissues with dysregulated receptors in the intratumoral CD4+ T cells were selected as the predictor of the IPSE of tamoxifen. The patients predicted to suffer low IPSE (low Th2 polarization) had a significantly lower distant relapse risk than the patients predicted to suffer high IPSE in independent datasets (n = 608; HR = 4.326, P = 0.000897; HR = 2.014, P = 0.0173; HR = 2.72, P = 0.04077). Patients predicted to suffer low IPSE would benefit from tamoxifen treatment (HR = 2.908, P = 0.03905). The DE intratumoral cytokines identified in this study may help predict the IPSE of tamoxifen and justify the use of tamoxifen in BC treatment.

Differences in the transcriptional response to fulvestrant and estrogen deprivation in ER-positive breast cancer

PURPOSE

Endocrine therapies include aromatase inhibitors and the selective estrogen receptor (ER) downregulator fulvestrant. This study aimed to determine whether the reported efficacy of fulvestrant over anastrozole, and high- over low-dose fulvestrant, reflect distinct transcriptional responses.

EXPERIMENTAL DESIGN

Global gene expression profiles from ERα-positive breast carcinomas before and during presurgical treatment with fulvestrant (n = 22) or anastrozole (n = 81), and corresponding in vitro models, were compared. Transcripts responding differently to fulvestrant and estrogen deprivation were identified and integrated using Gene Ontology, pathway and network analyses to evaluate their potential significance.

RESULTS

The overall transcriptional response to fulvestrant and estrogen deprivation was correlated (r = 0.61 in presurgical studies, r = 0.87 in vitro), involving downregulation of estrogen-regulated and proliferation-associated genes. The transcriptional response to fulvestrant was of greater magnitude than estrogen deprivation (slope = 0.62 in presurgical studies, slope = 0.63 in vitro). Comparative analyses identified 28 genes and 40 Gene Ontology categories affected differentially by fulvestrant. Seventeen fulvestrant-specific genes, including CAV1/2, SNAI2, and NRP1, associated with ERα, androgen receptor (AR), and TP53, in a network regulating cell cycle, death, survival, and tumor morphology. Eighteen genes responding differently to fulvestrant specifically predicted antiproliferative response to fulvestrant, but not anastrozole. Transcriptional effects of low-dose fulvestrant correlated with high-dose treatment, but were of lower magnitude (ratio = 0.29).

CONCLUSIONS

The transcriptional response to fulvestrant has much in common with estrogen deprivation, but is stronger with distinctions potentially attributable to arrest of estrogen-independent ERα activity and involvement of AR signaling. Genes responding differently to fulvestrant may have predictive utility. These data are consistent with the clinical efficacy of fulvestrant versus anastrozole and higher dosing regimens.

Next-generation biobanking of metastases to enable multidimensional molecular profiling in personalized medicine

Great advances in analytical technology coupled with accelerated new drug development and growing understanding of biological challenges, such as tumor heterogeneity, have required a change in the focus for biobanking. Most current banks contain samples of primary tumors, but linking molecular signatures to therapeutic questions requires serial biopsies in the setting of metastatic disease, next-generation of biobanking. Furthermore, an integration of multidimensional analysis of various molecular components, that is, RNA, DNA, methylome, microRNAome and post-translational modifications of the proteome, is necessary for a comprehensive view of a tumor's biology. While data using such biopsies are now regularly presented, the preanalytical variables in tissue procurement and processing in multicenter studies are seldom detailed and therefore are difficult to duplicate or standardize across sites and across studies. In the context of a biopsy-driven clinical trial, we generated a detailed protocol that includes morphological evaluation and isolation of high-quality nucleic acids from small needle core biopsies obtained from liver metastases. The protocol supports stable shipping of samples to a central laboratory, where biopsies are subsequently embedded in support media. Designated pathologists must evaluate all biopsies for tumor content and macrodissection can be performed if necessary to meet our criteria of >60% neoplastic cells and <20% necrosis for genomic isolation. We validated our protocol in 40 patients who participated in a biopsy-driven study of therapeutic resistance in metastatic colorectal cancer. To ensure that our protocol was compatible with multiplex discovery platforms and that no component of the processing interfered with downstream enzymatic reactions, we performed array comparative genomic hybridization, methylation profiling, microRNA profiling, splicing variant analysis and gene expression profiling using genomic material isolated from liver biopsy cores. Our standard operating procedures for next-generation biobanking can be applied widely in multiple settings, including multicentered and international biopsy-driven trials.

MMAD: microarray microdissection with analysis of differences is a computational tool for deconvoluting cell type-specific contributions from tissue samples

BACKGROUND

One of the significant obstacles in the development of clinically relevant microarray-derived biomarkers and classifiers is tissue heterogeneity. Physical cell separation techniques, such as cell sorting and laser-capture microdissection, can enrich samples for cell types of interest, but are costly, labor intensive and can limit investigation of important interactions between different cell types.

RESULTS

We developed a new computational approach, called microarray microdissection with analysis of differences (MMAD), which performs microdissection in silico. Notably, MMAD (i) allows for simultaneous estimation of cell fractions and gene expression profiles of contributing cell types, (ii) adjusts for microarray normalization bias, (iii) uses the corrected Akaike information criterion during model optimization to minimize overfitting and (iv) provides mechanisms for comparing gene expression and cell fractions between samples in different classes. Computational microdissection of simulated and experimental tissue mixture datasets showed tight correlations between predicted and measured gene expression of pure tissues as well as tight correlations between reported and estimated cell fraction for each of the individual cell types. In simulation studies, MMAD showed superior ability to detect differentially expressed genes in mixed tissue samples when compared with standard metrics, including both significance analysis of microarrays and cell type-specific significance analysis of microarrays.

CONCLUSIONS

We have developed a new computational tool called MMAD, which is capable of performing robust tissue microdissection in silico, and which can improve the detection of differentially expressed genes. MMAD software as implemented in MATLAB is publically available for download at http://sourceforge.net/projects/mmad/.

A self-directed method for cell-type identification and separation of gene expression microarrays

Gene expression analysis is generally performed on heterogeneous tissue samples consisting of multiple cell types. Current methods developed to separate heterogeneous gene expression rely on prior knowledge of the cell-type composition and/or signatures - these are not available in most public datasets. We present a novel method to identify the cell-type composition, signatures and proportions per sample without need for a-priori information. The method was successfully tested on controlled and semi-controlled datasets and performed as accurately as current methods that do require additional information. As such, this method enables the analysis of cell-type specific gene expression using existing large pools of publically available microarray datasets.

Gene expression microarrays are widely used to uncover biological insights. Most microarray experiments profile whole tissues containing mixtures of multiple cell-types. As such, gene expression differences between samples may be due to different cellular compositions or biological differences, highly limiting the conclusions derived from the analysis. All current approaches to computationally separate the heterogeneous gene expression to individual cell-types require that the identity, relative amount of the cell-types in the tissue or their individual gene expression are known. Publically available microarray-based datasets, which include thousands of patient samples, do not usually measure this information, rendering existing separation methods unusable. We developed a novel approach to estimate the number of cell-types, identities, individual gene expression and relative proportions in heterogeneous tissues with no a-priori information except for an initial estimate of the cell-types in the tissue analyzed and general reference signatures of these cell-types that may be easily obtained from public databases. We successfully applied our method to microarray datasets, yielding highly accurate estimations, which often exceed the performance of separation methods that require prior information. Thus, our method can be accurately applied to any heterogeneous dataset, where re-examination and analysis of the individual cell-types in the heterogeneous tissue can aid in discovering new aspects regarding these diseases.

Significance of TP53 mutations determined by next-generation "deep" sequencing in prognosis of estrogen receptor-positive breast cancer

Next-generation "deep" sequencing (NGS) was used for the mutational analysis of TP53, and a DNA microarray was used for the determination of the TP53 mutation-associated gene expression signature (TP53 GES) in 115 estrogen receptor (ER)-positive breast cancers. NGS detected 27 TP53 mutations, of which 20 were also detected by Sanger sequencing (SS) and seven were detected only by NGS. A significantly higher number of mutant alleles (33.9%) was detected in the tumors with TP53 mutations detected by SS compared with those with TP53 mutations detected only by NGS (11.1%). The TP53 mutations detected by NGS were more significantly associated with a large tumor size, a high histological grade, progesterone receptor-negativity, and HER2-positivity compared with those detected by SS. The TP53 mutations detected by SS, but not those detected by NGS or the p53 immunohistochemistry, exhibited a significant association with poor prognosis. In addition, the TP53 GES more clearly differentiated low- from high-risk patients for relapse than the TP53 mutations detected by SS, regardless of the other conventional prognostic factors. Thus, NGS is more sensitive for the detection of TP53 mutations, but the prognostic significance of these mutations could not be demonstrated. In contrast, the TP53 GES proved to be a powerful prognostic indicator for ER-positive tumors.

Thrombospondin-4 expression is activated during the stromal response to invasive breast cancer

The thromobospondins are a family of extracellular glycoproteins that are activated during tissue remodeling processes such as embryogenesis, wound healing and cancer. Thrombospondin-4 (THBS4) is known to have roles in cellular migration, adhesion and attachment, as well as proliferation in different contexts. Data to support a role in cancer biology is increasing, including for gastrointestinal and prostate tumours. Here, using a combination of immunohistochemistry, immunofluorescence and analysis of publicly available genomic and expression data, we present the first study describing the pattern of expression of THBS4 in normal breast and breast cancer. THBS4 was located to the basement membrane of large ducts and vessels in normal breast tissue, but was absent from epithelium and extracellular matrix. There was a significant induction in expression in cancer-associated stroma relative to normal stroma (P = 0.0033), neoplastic epithelium (P < 0.0001) and normal epithelium (P < 0.0001). There was no difference in stromal expression of THBS4 between invasive ductal carcinomas (IDC) and invasive lobular carcinomas (ILC). The THBS4 mRNA levels were variable yet were generally highest in tumours typically rich in stromal content (ILC, ER positive low grade IDC; luminal A and normal-like subtypes). Genomic alterations of the THBS4 gene (somatic mutations and gene copy number) are rare suggesting this dramatic activation in expression is most likely dynamically regulated through the interaction between invading tumour cells and stromal fibroblasts in the local microenvironment. In summary, THBS4 expression in breast cancer-associated extracellular matrix contributes to the activated stromal response exhibited during tumour progression and this may facilitate invasion of tumour cells.

Characteristics of cross-hybridization and cross-alignment of expression in pseudo-xenograft samples by RNA-Seq and microarrays

Background

Exploring stromal changes associated with tumor growth and development is a growing area of oncologic research. In order to study molecular changes in the stroma it is recommended to separate tumor tissue from stromal tissue. This is relevant to xenograft models where tumors can be small and difficult to separate from host tissue. We introduce a novel definition of cross-alignment/cross-hybridization to compare qualitatively the ability of high-throughput mRNA sequencing, RNA-Seq, and microarrays to detect tumor and stromal expression from mixed ‘pseudo-xenograft’ samples vis-à-vis genes and pathways in cross-alignment (RNA-Seq) and cross-hybridization (microarrays). Samples consisted of normal mouse lung and human breast cancer cells; these were combined in fixed proportions to create a titration series of 25% steps. Our definition identifies genes in a given species (human or mouse) with undetectable expression in same-species RNA but detectable expression in cross-species RNA. We demonstrate the comparative value of this method and discuss its potential contribution in cancer research.

Results

Our method can identify genes from either species that demonstrate cross-hybridization and/or cross-alignment properties. Surprisingly, the set of genes identified using a simpler and more common approach (using a ‘pure’ cross-species sample and calling all detected genes as ‘crossers’) is not a superset of the genes identified using our technique. The observed levels of cross-hybridization are relatively low: 5.3% of human genes detected in mouse, and 3.5% of mouse genes detected in human. Observed levels of cross-alignment are practically comparable to the levels of cross-hybridization: 6.5% of human genes detected in mouse, and 2.3% of mouse genes detected in human. We also observed a relatively high percentage of orthologs: 40.3% of cross-hybridizing genes, and 32.2% of cross-aligning genes.

Normalizing the gene catalog to use Consensus Coding Sequence (CCDS) IDs (Genome Res 19:1316–1323, 2009), our results show that the observed levels of cross-hybridization are low: 2.7% of human CCDS IDs are detected in mouse, and 2.4% of mouse CCDS IDs are detected in human. Levels of cross-alignment using the RNA-Seq data are comparable for the mouse, 2.2% of mouse CCDS IDs detected in human, and 9.9% of human CCDS IDs detected in mouse. However, the lists of cross-aligning/cross-hybridizing genes contain many that are of specific interest to oncologic researchers.

Conclusions

The conservative definition that we propose identifies genes in mouse whose expression can be attributed to human RNA, and vice versa, as well as revealing genes with cross-alignment/cross-hybridization properties which could not be identified using a simpler but more established approach. The overall percentage of genes affected by cross-hybridization/cross-alignment is small, but includes genes that are of interest to oncologic researchers. Which platform to use with mixed xenograft samples, microarrays or RNA-Seq, appears to be primarily a question of cost and whether the detection and measurement of expression of specific genes of interest are likely to be affected by cross-hybridization or cross-alignment.

Integrative molecular and functional profiling of ERBB2-amplified breast cancers identifies new genetic dependencies

Overexpression of the receptor tyrosine kinase ERBB2 (also known as HER2) occurs in around 15% of breast cancers and is driven by amplification of the ERBB2 gene. ERBB2 amplification is a marker of poor prognosis, and although anti-ERBB2-targeted therapies have shown significant clinical benefit, de novo and acquired resistance remains an important problem. Genomic profiling has demonstrated that ERBB2+ve breast cancers are distinguished from ER+ve and 'triple-negative' breast cancers by harbouring not only the ERBB2 amplification on 17q12, but also a number of co-amplified genes on 17q12 and amplification events on other chromosomes. Some of these genes may have important roles in influencing clinical outcome, and could represent genetic dependencies in ERBB2+ve cancers and therefore potential therapeutic targets. Here, we describe an integrated genomic, gene expression and functional analysis to determine whether the genes present within amplicons are critical for the survival of ERBB2+ve breast tumour cells. We show that only a fraction of the ERBB2-amplified breast tumour lines are truly addicted to the ERBB2 oncogene at the mRNA level and display a heterogeneous set of additional genetic dependencies. These include an addiction to the transcription factor gene TFAP2C when it is amplified and overexpressed, suggesting that TFAP2C represents a genetic dependency in some ERBB2+ve breast cancer cells.

Molecular pathology of breast cancer: what a pathologist needs to know

Pathologists are now more than ever "diagnostic oncologists" and serve a critical role as clinical consultants on the biology of disease. In the last decade and a half, molecular information has transformed our thinking about the biologic diversity of breast cancers and redirected the way clinical treatment decisions are made. A basic understanding of the current molecular classification of breast cancers and the biologic pathways from precursors to invasive disease is key to both informing diagnostic practice and serving as clinical consultants. In addition, both single-marker and panel-based molecular tests are currently being utilized in breast cancer tissue to predict the benefit of specific therapies such as HER2-targeted biologic therapy and chemotherapy. Familiarity with the current issues involving these molecular tests as well as the pathologist's role in ensuring appropriate tissue handling, tissue selection, and results interpretation and correlation are paramount to providing optimal patient care.

Molecular biology in breast cancer: should molecular classifiers be assessed by conventional tools or by gene expression arrays?

Breast cancer is a complex disease, with heterogeneous presentations and clinical courses. Standard clinico-pathological parameters, relying on single gene or protein characterization determined with sometimes poorly-reproducible technologies, have shown limitations in the classification of the disease and in the prediction of individual patient outcomes and responses to therapy. Gene-expression profiling has revealed great potential to accurately classify breast cancer and define patient prognosis and prediction to anti-cancer therapy. Nevertheless, the performance of molecular classifiers remains sub-optimal, and both technical and conceptual improvements are needed. It is likely that determining the ideal strategy for tailoring treatment of breast cancer will require a more systematic, structured and multi-dimensional approach than in the past. Besides implementing cutting-edge technologies to detect genetic and epigenetic cancer alterations, the future of breast cancer research will in all probability rely on the innovative and multilevel integration of molecular profiles with clinical parameters of the disease and patient-related factors.

The renin-angiotensin system in the breast and breast cancer

Much evidence now suggests that angiotensin II has roles in normal functions of the breast that may be altered or attenuated in cancer. Both angiotensin type 1 (AT1) and type 2 (AT2) receptors are present particularly in the secretory epithelium. Additionally, all the elements of a tissue renin-angiotensin system, angiotensinogen, prorenin and angiotensin-converting enzyme (ACE), are also present and distributed in different cell types in a manner suggesting a close relationship with sites of angiotensin II activity. These findings are consistent with the concept that stromal elements and myoepithelium are instrumental in maintaining normal epithelial structure and function. In disease, this system becomes disrupted, particularly in invasive carcinoma. Both AT1 and AT2 receptors are present in tumours and may be up-regulated in some. Experimentally, angiotensin II, acting via the AT1 receptor, increases tumour cell proliferation and angiogenesis, both these are inhibited by blocking its production or function. Epidemiological evidence on the effect of expression levels of ACE or the distribution of ACE or AT1 receptor variants in many types of cancer gives indirect support to these concepts. It is possible that there is a case for the therapeutic use of high doses of ACE inhibitors and AT1 receptor blockers in breast cancer, as there may be for AT2 receptor agonists, though this awaits full investigation. Attention is drawn to the possibility of blocking specific AT1-mediated intracellular signalling pathways, for example by AT1-directed antibodies, which exploit the possibility that the extracellular N-terminus of the AT1 receptor may have previously unsuspected signalling roles.

Molecular classification of estrogen receptor-positive/luminal breast cancers

Estrogen receptor (ER)-positive breast cancer is the most prevalent subtype of invasive breast cancers. Patients with ER-positive breast cancers have variable clinical outcomes and responses to endocrine therapy and chemotherapy. With the advent of microarray-based gene expression profiling, unsupervised analysis methods have resulted in a classification of ER-positive disease into subtypes with different outcomes (ie, luminal A and luminal B); subsequent studies have demonstrated that these subtypes have different patterns of genetic aberrations and outcome. Studies based on supervised methods of microarray analysis have led to the development of prognostic gene signatures that identify a subgroup of ER-positive breast cancer patients with excellent outcome, who could forego chemotherapy. Despite the excitement with these approaches, several lines of evidence have demonstrated that the subclassification of ER-positive cancers and the prognostic value of gene signatures is largely driven by the expression levels of proliferation-related genes and that proliferation markers, such as Ki67, may provide equivalent prognostic information to that provided by gene signatures. In this review, we discuss the contribution of gene expression profiling to the classification of ER-positive breast cancer, the role of prognostic and predictive signatures, and the potential stratification of ER-positive disease according to their dependency on the phosphatidylinositol 3-kinase pathway.

Semi-supervised Nonnegative Matrix Factorization for gene expression deconvolution: a case study

Heterogeneity in sample composition is an inherent issue in many gene expression studies and, in many cases, should be taken into account in the downstream analysis to enable correct interpretation of the underlying biological processes. Typical examples are infectious diseases or immunology-related studies using blood samples, where, for example, the proportions of lymphocyte sub-populations are expected to vary between cases and controls. Nonnegative Matrix Factorization (NMF) is an unsupervised learning technique that has been applied successfully in several fields, notably in bioinformatics where its ability to extract meaningful information from high-dimensional data such as gene expression microarrays has been demonstrated. Very recently, it has been applied to biomarker discovery and gene expression deconvolution in heterogeneous tissue samples. Being essentially unsupervised, standard NMF methods are not guaranteed to find components corresponding to the cell types of interest in the sample, which may jeopardize the correct estimation of cell proportions. We have investigated the use of prior knowledge, in the form of a set of marker genes, to improve gene expression deconvolution with NMF algorithms. We found that this improves the consistency with which both cell type proportions and cell type gene expression signatures are estimated. The proposed method was tested on a microarray dataset consisting of pure cell types mixed in known proportions. Pearson correlation coefficients between true and estimated cell type proportions improved substantially (typically from about 0.5 to approximately 0.8) with the semi-supervised (marker-guided) versions of commonly used NMF algorithms. Furthermore known marker genes associated with each cell type were assigned to the correct cell type more frequently for the guided versions. We conclude that the use of marker genes improves the accuracy of gene expression deconvolution using NMF and suggest modifications to how the marker gene information is used that may lead to further improvements.

Challenges translating breast cancer gene signatures into the clinic

The advent of microarray-based gene-expression profiling a decade ago raised high expectations for rapid advances in breast cancer classification, prognostication and prediction. Despite the development of molecular classifications, and prognostic and predictive gene-expression signatures, microarray-based studies have not yielded definitive answers to many of the questions that remain germane for the successful implementation of personalized medicine. There are a lack of robust signatures to predict benefit from specific therapeutic agents and it is still not possible to predict prognosis or chemotherapy treatment response in specific disease subsets accurately, such as triple-negative breast cancer. We discuss the hurdles in the development and validation of molecular classification systems, and prognostic and predictive signatures based on microarray gene-expression profiling. We suggest that similar challenges are likely to be encountered in translating next-generation sequencing data into clinically useful information. Finally we highlight strategies for the development of clinically useful molecular predictors in the future.

Systematic bias in genomic classification due to contaminating non-neoplastic tissue in breast tumor samples

Background

Genomic tests are available to predict breast cancer recurrence and to guide clinical decision making. These predictors provide recurrence risk scores along with a measure of uncertainty, usually a confidence interval. The confidence interval conveys random error and not systematic bias. Standard tumor sampling methods make this problematic, as it is common to have a substantial proportion (typically 30-50%) of a tumor sample comprised of histologically benign tissue. This "normal" tissue could represent a source of non-random error or systematic bias in genomic classification.

Methods

To assess the performance characteristics of genomic classification to systematic error from normal contamination, we collected 55 tumor samples and paired tumor-adjacent normal tissue. Using genomic signatures from the tumor and paired normal, we evaluated how increasing normal contamination altered recurrence risk scores for various genomic predictors.

Results

Simulations of normal tissue contamination caused misclassification of tumors in all predictors evaluated, but different breast cancer predictors showed different types of vulnerability to normal tissue bias. While two predictors had unpredictable direction of bias (either higher or lower risk of relapse resulted from normal contamination), one signature showed predictable direction of normal tissue effects. Due to this predictable direction of effect, this signature (the PAM50) was adjusted for normal tissue contamination and these corrections improved sensitivity and negative predictive value. For all three assays quality control standards and/or appropriate bias adjustment strategies can be used to improve assay reliability.

Conclusions

Normal tissue sampled concurrently with tumor is an important source of bias in breast genomic predictors. All genomic predictors show some sensitivity to normal tissue contamination and ideal strategies for mitigating this bias vary depending upon the particular genes and computational methods used in the predictor.

Microarrays in the 2010s: the contribution of microarray-based gene expression profiling to breast cancer classification, prognostication and prediction

Breast cancer comprises a collection of diseases with distinctive clinical, histopathological, and molecular features. Importantly, tumors with similar histological features may display disparate clinical behaviors. Gene expression profiling using microarray technologies has improved our understanding of breast cancer biology and has led to the development of a breast cancer molecular taxonomy and of multigene 'signatures' to predict outcome and response to systemic therapies. The use of these prognostic and predictive signatures in routine clinical decision-making remains controversial. Here, we review the clinical relevance of microarray-based profiling of breast cancer and discuss its impact on patient management.

Integrative computational biology for cancer research

Over the past two decades, high-throughput (HTP) technologies such as microarrays and mass spectrometry have fundamentally changed clinical cancer research. They have revealed novel molecular markers of cancer subtypes, metastasis, and drug sensitivity and resistance. Some have been translated into the clinic as tools for early disease diagnosis, prognosis, and individualized treatment and response monitoring. Despite these successes, many challenges remain: HTP platforms are often noisy and suffer from false positives and false negatives; optimal analysis and successful validation require complex workflows; and great volumes of data are accumulating at a rapid pace. Here we discuss these challenges, and show how integrative computational biology can help diminish them by creating new software tools, analytical methods, and data standards.

Revisiting the technical validation of tumour biomarker assays: how to open a Pandora's box

A tumour biomarker is a characteristic that is objectively measured and evaluated in tumour samples as an indicator of normal biological processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention. The development of a biomarker contemplates distinct phases, including discovery by hypothesis-generating preclinical or exploratory studies, development and qualification of the assay for the identification of the biomarker in clinical samples, and validation of its clinical significance. Although guidelines for the development and validation of biomarkers are available, their implementation is challenging, owing to the diversity of biomarkers being developed. The term 'validation' undoubtedly has several meanings; however, in the context of biomarker research, a test may be considered valid if it is 'fit for purpose'. In the process of validation of a biomarker assay, a key point is the validation of the methodology. Here we discuss the challenges for the technical validation of immunohistochemical and gene expression assays to detect tumour biomarkers and provide suggestions of pragmatic solutions to address these challenges.

Impact of sample heterogeneity on methylation analysis

The recent emergence of high-throughput arrays for methylation analysis has made the influence of tumor content on the interpretation of methylation levels increasingly pertinent. However, to what degree does tumor content have an influence, and what degree of tumor content makes a specimen acceptable for accurate analysis remains unclear. Taking a systematic approach, we analyzed 98 unselected formalin-fixed and paraffin-embedded gastric tumors and matched normal tissue samples using the Illumina GoldenGate methylation assay. Unsupervised hierarchical clustering showed 2 separate clusters with a significant difference in average tumor content levels. The probes identified to be significantly differentially methylated between the tumors and normals also differed according to the tumor content of the samples included, with the sensitivity of identifying the "top" candidate probes significantly reduced when including samples below 70% tumor content. We also tested whether the removal of the probes featuring single nucleotide polymorphisms and/or DNA repetitive elements, reportedly present in GoldenGate arrays, would significantly affect the study's findings, and found little change in the results with their omission. Our findings suggest that tumor content significantly influences the interpretation of methylation levels and candidate gene identification, and that 70% tumor content may be a suitable threshold for selecting samples for methylation studies.

Gene expression deconvolution in clinical samples

Cell type heterogeneity may have a substantial effect on gene expression profiling of human tissue. Several in silico methods for deconvoluting a gene expression profile into cell-type-specific subprofiles have been published but not widely used. Here, we consider recent methods and the experimental validations available for them. Shen-Orr et al. recently developed an approach called cell-type-specific significance analysis of microarray for deconvoluting gene expression. This method requires the measurement of the proportion of each cell type in each sample and the expression profiles of the heterogeneous samples. It determines how gene expression varies among pre-defined phenotypes for each cell type. Gene expression can vary substantially among cell types and sample heterogeneity can mask the identification of biologically important phenotypic correlations. Consequently, the deconvolution approach can be useful in the analysis of mixtures of cell populations in clinical samples.

Molecular profiling: moving away from tumor philately

Despite the remarkable enhancements in breast cancer classification and characterization, an exhaustive comprehension of the underlying carcinogenic processes and an accurate prediction of its clinical behavior are yet to be achieved. On the wave of recent scientific advances and clinical findings, new cutting-edge technologies are poised to deliver an array of diverse molecular data that are expected to dramatically alter breast cancer diagnosis and treatment.

In silico ascription of gene expression differences to tumor and stromal cells in a model to study impact on breast cancer outcome

Breast tumors consist of several different tissue components. Despite the heterogeneity, most gene expression analyses have traditionally been performed without prior microdissection of the tissue sample. Thus, the gene expression profiles obtained reflect the mRNA contribution from the various tissue components. We utilized histopathological estimations of area fractions of tumor and stromal tissue components in 198 fresh-frozen breast tumor tissue samples for a cell type-associated gene expression analysis associated with distant metastasis. Sets of differentially expressed gene-probes were identified in tumors from patients who developed distant metastasis compared with those who did not, by weighing the contribution from each tumor with the relative content of stromal and tumor epithelial cells in their individual tumor specimen. The analyses were performed under various assumptions of mRNA transcription level from tumor epithelial cells compared with stromal cells. A set of 30 differentially expressed gene-probes was ascribed solely to carcinoma cells. Furthermore, two sets of 38 and five differentially expressed gene-probes were mostly associated to tumor epithelial and stromal cells, respectively. Finally, a set of 26 differentially expressed gene-probes was identified independently of cell type focus. The differentially expressed genes were validated in independent gene expression data from a set of laser capture microdissected invasive ductal carcinomas. We present a method for identifying and ascribing differentially expressed genes to tumor epithelial and/or stromal cells, by utilizing pathologic information and weighted t-statistics. Although a transcriptional contribution from the stromal cell fraction is detectable in microarray experiments performed on bulk tumor, the gene expression differences between the distant metastasis and no distant metastasis group were mostly ascribed to the tumor epithelial cells of the primary breast tumors. However, the gene PIP5K2A was found significantly elevated in stroma cells in distant metastasis group, compared to stroma in no distant metastasis group. These findings were confirmed in gene expression data from the representative compartments from microdissected breast tissue. The method described was also found to be robust to different histopathological procedures.

In silico estimates of tissue components in surgical samples based on expression profiling data

Tissue samples from many diseases have been used for gene expression profiling studies, but these samples often vary widely in the cell types they contain. Such variation could confound efforts to correlate expression with clinical parameters. In principle, the proportion of each major tissue component can be estimated from the profiling data and used to triage samples before studying correlations with disease parameters. Four large gene expression microarray data sets from prostate cancer, whose tissue components were estimated by pathologists, were used to test the performance of multivariate linear regression models for in silico prediction of major tissue components. Ten-fold cross-validation within each data set yielded average differences between the pathologists' predictions and the in silico predictions of 8% to 14% for the tumor component and 13% to 17% for the stroma component. Across independent data sets that used similar platforms and fresh frozen samples, the average differences were 11% to 12% for tumor and 12% to 17% for stroma. When the models were applied to 219 arrays of "tumor-enriched" samples in the literature, almost one quarter were predicted to have 30% or less tumor cells. Furthermore, there was a 10.5% difference in the average predicted tumor content between 37 recurrent and 42 nonrecurrent cancer patients. As a result, genes that correlated with tissue percentage generally also correlated with recurrence. If such a correlation is not desired, then some samples might be removed to rebalance the data set or tissue percentages might be incorporated into the prediction algorithm. A web service, "CellPred," has been designed for the in silico prediction of sample tissue components based on expression data.

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.

Statistical expression deconvolution from mixed tissue samples

MOTIVATION

Global expression patterns within cells are used for purposes ranging from the identification of disease biomarkers to basic understanding of cellular processes. Unfortunately, tissue samples used in cancer studies are usually composed of multiple cell types and the non-cancerous portions can significantly affect expression profiles. This severely limits the conclusions that can be made about the specificity of gene expression in the cell-type of interest. However, statistical analysis can be used to identify differentially expressed genes that are related to the biological question being studied.

RESULTS

We propose a statistical approach to expression deconvolution from mixed tissue samples in which the proportion of each component cell type is unknown. Our method estimates the proportion of each component in a mixed tissue sample; this estimate can be used to provide estimates of gene expression from each component. We demonstrate our technique on xenograft samples from breast cancer research and publicly available experimental datasets found in the National Center for Biotechnology Information Gene Expression Omnibus repository.

AVAILABILITY

R code (http://www.r-project.org/) for estimating sample proportions is freely available to non-commercial users and available at http://www.med.miami.edu/medicine/x2691.xml.

Breast cancer molecular profiling with single sample predictors: a retrospective analysis

BACKGROUND

Microarray expression profiling classifies breast cancer into five molecular subtypes: luminal A, luminal B, basal-like, HER2, and normal breast-like. Three microarray-based single sample predictors (SSPs) have been used to define molecular classification of individual samples. We aimed to establish agreement between these SSPs for identification of breast cancer molecular subtypes.

METHODS

Previously described microarray-based SSPs were applied to one in-house (n=53) and three publicly available (n=779) breast cancer datasets. Agreement was analysed between SSPs for the whole classification system and for the five molecular subtypes individually in each cohort.

FINDINGS

Fair-to-substantial agreement between every pair of SSPs in each cohort was recorded (kappa=0.238-0.740). Of the five molecular subtypes, only basal-like cancers consistently showed almost-perfect agreement (kappa>0.812). The proportion of cases classified as basal-like in each cohort was consistent irrespective of the SSP used; however, the proportion of each remaining molecular subtype varied substantially. Assignment of individual cases to luminal A, luminal B, HER2, and normal breast-like subtypes was dependent on the SSP used. The significance of associations with outcome of each molecular subtype, other than basal-like and luminal A, varied depending on SSP used. However, different SSPs produced broadly similar survival curves.

INTERPRETATION

Although every SSP identifies molecular subtypes with similar survival, they do not reliably assign the same patients to the same molecular subtypes. For molecular subtype classification to be incorporated into routine clinical practice and treatment decision making, stringent standardisation of methodologies and definitions for identification of breast cancer molecular subtypes is needed.

FUNDING

Breakthrough Breast Cancer, Cancer Research UK.

PSMB7 is associated with anthracycline resistance and is a prognostic biomarker in breast cancer

BACKGROUND

To date individual markers have failed to correctly predict resistance against anticancer agents in breast cancer. We used gene expression patterns attributable to chemotherapy-resistant cells to detect potential new biomarkers related to anthracycline resistance. One of the genes, PSMB7, was selected for further functional studies and clinical validation.

METHODS

We contrasted the expression profiles of four pairs of different human tumour cell lines and of their counterparts resistant to doxorubicin. Observed overexpression of PSMB7 in resistant cell lines was validated by immunohistochemistry. To examine its function in chemoresistance, we silenced the gene by RNA interference (RNAi) in doxorubicin-resistant MCF-7 breast cancer cells, then cell vitality was measured after doxorubicin treatment. Microarray gene expression from GEO raw microarray samples with available progression-free survival data was downloaded, and expression of PSMB7 was used for grouping samples.

RESULTS

After doxorubicin treatment, 79.8+/-13.3% of resistant cells survived. Silencing of PSMB7 in resistant cells decreased survival to 31.8+/-6.4% (P>0.001). A similar effect was observed after paclitaxel treatment. In 1592 microarray samples, the patients with high PSMB7 expression had a significantly shorter survival than the patients with low expression (P<0.001).

CONCLUSION

Our findings suggest that high PSMB7 expression is an unfavourable prognostic marker in breast cancer.

Microarray-based Gene Expression Profiling as a Clinical Tool for Breast Cancer Management: Are We There Yet?

Breast cancer is a heterogeneous disease, encompassing several histological types and clinical behaviors. Current histopathological classification systems are based on descriptive entities with prognostic significance. Few prognostic and predictive markers beyond those offered by histopathological analysis are available. High-throughput molecular technologies are reshaping our understanding of breast cancer, of which microarray-based gene expression has received most attention. This method has been used to derive a molecular taxonomy for breast cancer, which has provided interesting insights into the biology of the disease. Class prediction studies have generated a multitude of prognostic/predictive signatures, which herald the promise for an improvement in treatment decision making. However, most of the signatures developed to date seem to have discriminatory power almost restricted to estrogen receptor—positive disease. This review addresses the contribution of gene expression profiling to our understanding of breast cancer and its clinical management.

Biomarkers in oncology drug development

Biomarker measurements have become an essential component of oncology drug development, particularly so in this era of targeted therapies. Such measurements ensure that clinical studies are testing our biological hypotheses and can help make the difficult decisions required to choose which drugs to stop developing or de-prioritise. For those drugs taken forward, biomarker measurements may also help choose the appropriate dose, schedule and patient population. In this review we discuss the intrinsic properties of biological sample based efficacy measurements and how these relate to their implementation in oncology drug development by way of points to consider and examples.

Evaluation of the current knowledge limitations in breast cancer research: a gap analysis

BACKGROUND

A gap analysis was conducted to determine which areas of breast cancer research, if targeted by researchers and funding bodies, could produce the greatest impact on patients.

METHODS

Fifty-six Breast Cancer Campaign grant holders and prominent UK breast cancer researchers participated in a gap analysis of current breast cancer research. Before, during and following the meeting, groups in seven key research areas participated in cycles of presentation, literature review and discussion. Summary papers were prepared by each group and collated into this position paper highlighting the research gaps, with recommendations for action.

RESULTS

Gaps were identified in all seven themes. General barriers to progress were lack of financial and practical resources, and poor collaboration between disciplines. Critical gaps in each theme included: (1) genetics (knowledge of genetic changes, their effects and interactions); (2) initiation of breast cancer (how developmental signalling pathways cause ductal elongation and branching at the cellular level and influence stem cell dynamics, and how their disruption initiates tumour formation); (3) progression of breast cancer (deciphering the intracellular and extracellular regulators of early progression, tumour growth, angiogenesis and metastasis); (4) therapies and targets (understanding who develops advanced disease); (5) disease markers (incorporating intelligent trial design into all studies to ensure new treatments are tested in patient groups stratified using biomarkers); (6) prevention (strategies to prevent oestrogen-receptor negative tumours and the long-term effects of chemoprevention for oestrogen-receptor positive tumours); (7) psychosocial aspects of cancer (the use of appropriate psychosocial interventions, and the personal impact of all stages of the disease among patients from a range of ethnic and demographic backgrounds).

CONCLUSION

Through recommendations to address these gaps with future research, the long-term benefits to patients will include: better estimation of risk in families with breast cancer and strategies to reduce risk; better prediction of drug response and patient prognosis; improved tailoring of treatments to patient subgroups and development of new therapeutic approaches; earlier initiation of treatment; more effective use of resources for screening populations; and an enhanced experience for people with or at risk of breast cancer and their families. The challenge to funding bodies and researchers in all disciplines is to focus on these gaps and to drive advances in knowledge into improvements in patient care.

The oestrogen-dependent biology of breast cancer. Sensitivity and resistance to aromatase inhibitors revisited: a molecular perspective

Endocrine treatment of breast cancer was the first molecular targeted anti-cancer therapy to reach clinical practice. Among the several options that share the common denomination of hormonal treatment, aromatase inhibitors (AIs) in the postmenopausal setting show the highest efficacy rates. These drugs have become the standard of care both in the advanced and adjuvant scenarios. Nevertheless resistance to AIs either upfront or after initial clinical response is almost a universal feature whenever tumour excision is not possible. Multiple reports have established the role of alternative pro-growth signalling pathways in the acquisition of resistance to the oestradiol deprivation that AIs produce. However the first clinical trials addressing the double blockade of both the oestrogen and other growing factor pathways raise some concerns on the efficacy of this approach. This review presents the evidence on the molecular events underpinning the response and resistance to AIs and suggests some key issues to consider when designing clinical research projects in this context.

Microarray-based prediction of tumor response to neoadjuvant radiochemotherapy of patients with locally advanced rectal cancer

BACKGROUND & AIMS

Neoadjuvant chemoradiotherapy has become a standard treatment of locally advanced rectal carcinomas, even though the responsiveness varies from complete response to resistance. The aim of the study was to evaluate the capacity of gene expression signatures to identify responders and nonresponders pretherapeutically.

METHODS

By using microarray technology we generated gene expression profiles of 43 biopsy specimens of locally advanced rectal carcinomas. The transcription profile then was compared with histopathologic response and used to identify a set of genes discriminating responders from nonresponders.

RESULTS

We identified a gene expression signature of 42 genes, mostly encoding proteins that either play a role in the nucleus, such as the transcription factor ETS2, or are associated with transport function, such as the solute carrier SLC35E1, or the regulation of apoptosis, such as caspase-1. In leave-one-out cross-validation the correct classification of a responder was 71%, the specificity of the analysis for a correct classification of a nonresponder was 86%. By applying an additional statistical method of 200 successive splittings into training and test data sets we generated an individual prediction accuracy measure for each predicted response.

CONCLUSIONS

Our study shows that pretherapeutic prediction of response of rectal carcinomas to neoadjuvant chemoradiotherapy is feasible, and may represent a new valuable and practical tool of therapeutic stratification.