Dataset Information

Breast tumor subtypes correlate with prognosis

ABSTRACT: To advance in our understanding of the biological pathways involved in breast cancer tumor progression we have analyzed a set of breast tumor biopsies in order to identify the genomic pathways in which tumor may develop. With this objective, a cDNA microarray platform containing 800 genes was constructed. These genes were chosen because they are in several representatives signaling pathways, namely estrogen and progesterone receptor related pathways, cell cycle, DNA repair, chromatin remodeling, cell proliferation, apoptosis, cell adhesion, cell invasion and angiogenesis. An analysis of the gene expression profile of a large group of breast tumors was carried out and good correlation with their clinical-histopathological data was obtained. We identified different tumor subclusters with distinctive phenotype characteristics within our population, which later correlated with clinical outcome. The most significant genes able to discriminate between different tumor phenotypes of our training set of samples were determined applying a “Leave-one-out” cross-validation method of statistical analysis called PAM (Prediction analysis of microarrays). The predictor was further tested on a new incoming set of samples where we determined to which subtype of tumor new samples were allocated and predicted outcome was compared to clinical data, showing how some distinct tumor phenotypes correlate with a poor prognosis. Pathway analysis of the most significant genes belonging to each phenotype was performed to elucidate the most representative biological signaling pathways through which tumor progression might elapse. We constructed a real time expression profiling platform with a selection of the most differentially expressed genes in each subtype, which could be used as a technique to help improve the diagnosis and prognosis of the breast tumor samples of our sample population.

ORGANISM(S): Homo sapiens

PROVIDER: GSE18908 | GEO | 2017/12/31

SECONDARY ACCESSION(S): PRJNA121321

REPOSITORIES: GEO

ACCESS DATA

Similar Datasets

Project description:Breast cancer is an extremely heterogeneous disease. This heterogeneity can be observed at multiple levels, including gene expression, chromosomal aberrations, and disease pathology. A clear understanding of these differences is important since they impact upon treatment efficacy and clinical outcome. Many studies have shown that the tumor microenvironment also plays a critical role in cancer initiation and progression. Although genomic technologies have been used to gain a better understanding of the impact of gene expression heterogeneity on breast cancer outcome by identifying gene expression signatures associated with clinical outcome, histopathological breast cancer subtypes, and a variety of cancer--related pathways and processes, relatively little is known about the influence of of heterogeneity in the tumor microenvironment on these factors. We show that gene expression in the breast tumor microenvironment is highly heterogeneous, identifying at least six different classes of tumor stroma with distinct expression patterns and distinct biological processes. Two of these classes recapitulate the processes identified in the stroma--derived prognostic predictor, while the others are new classes of stroma associated with distinct clinical outcomes. One of these is associated with matrix remodeling and is strongly associated with the basal molecular subtype of breast cancer. The remainder are independent of the previously published molecular subtypes of breast cancer. Additionally, based on independent data from over 800 tumors, the combinations of stroma classes and breast cancer subtypes identify new subgroups of breast tumors that show better discrimination between good and poor outcome individuals than the molecular breast cancer subtypes or the stroma classes alone, suggesting a novel classification scheme for breast cancer . This further demonstrates an important role for the tumor microenvironment in defining breast cancer heterogeneity, with a consequent impact upon clinical outcome.

Project description:Purpose HER2 gene amplification or protein overexpression (HER2+) defines a clinically challenging subgroup of breast cancer with variable prognosis and response to therapy. We aimed to investigate the heterogeneous biological appearance and clinical behavior of HER2+ tumors using molecular profiling. Materials and Methods Hierarchical clustering of gene expression data from 58 HER2-amplified tumors of various stage, histological grade and estrogen receptor (ER) status was used to construct a HER2-derived prognostic predictor that was further evaluated in several large independent breast cancer data sets. Results Unsupervised analysis identified three subtypes of HER2+ tumors with mixed stage, histological grade and ER-status. One subtype had a significantly worse clinical outcome. A prognostic predictor was created based on differentially expressed genes between the subtype with worse outcome and the other subtypes. The predictor was able to define patient groups with better and worse outcome in HER2+ breast cancer across multiple independent breast cancer data sets and identify a sizable HER2+ group with long disease-free survival and low mortality. Significant correlation to prognosis was also observed in basal-like, ER−, lymph node positive or high-grade tumors, irrespective of HER2-status. The predictor included genes associated to immune response, tumor invasion and metastasis. Conclusion The HER2-derived prognostic predictor provides further insight into the heterogeneous biology of HER2+ tumors and may become useful for improved selection of patients that need additional treatment with new drugs targeting the HER2 pathway. Array comparative genomic hybridization (aCGH) identified 58 breast tumors with amplification of HER2 from a larger cohort of approx 500 tumors breast. Global gene expression profiles were obtained using 70-mer oligonucleotide microarrays. Unsupervised hierarchical clustering of the 58 tumors, using Pearson correlation and complete linkage, identified three main clusters. One cluster showed significantly poorer clinical outcome. Significance of microarray (SAM) analysis was performed to identify 158 genes separating the poor outcome cluster compared to the other two clusters. Gene expression centroids, based on the 158 genes, were created for each cluster for validation in independent breast cancer data sets.

Project description:Accurate characterization and understanding of the breast cancer subtypes is of crucial clinical importance to the heterogeneity of this disease. Several layers of information, including immunohistochemical markers, mRNA and microRNA expression profiles, and pathway analysis have been used for such purpose in several studies. However, a comprehensively integrative approach is currently missing. This paper provides microRNA and mRNA expression profiles, characterizing four breast tumor subtypes, as defined by four immunohistochemical markers. The defined sets of features were validated in two independent data sets at multiple levels, including unsupervised clustering and supervised classification. Moreover, the gene expression signatures of the tumor subtypes were screened by in-depth analysis of 12 cancer core pathways. We successfully identified and validated a novel breast cancer subtypes gene expression signature composed of 976 mRNAs and 69 miRNAs. Luminal and non-luminal tumors are shown to significantly differ both at the mRNA and miRNA levels. HER2 positive tumors are more closely related to triple negative tumors by mRNA profiling than by miRNA expression. Closely related miRNAs sharing the same targets may exert opposite roles during tumor progression. Besides the core cancer pathways, other pathways such as those controling biomass synthesis are shown to be important to enable the core basal subtype with additional progressive nature compared with the other triple negative tumors. Some therapeutic strategies are proposed for breast cancer treatment, including the combined blockage of MAPK/ERK and PI3K/Pten signalings for tumors with poor clinical outcome, and targeting Wnt and JAK/STAT and/or Hedgehog, depending on tumor subtypes, together with conventional chemotherapy with the purpose of achieving an eradicative outcome. The pathway analysis also reveals that the clinical strategy and pivotal targets need to be tuned according to different tumor subtypes. This study is the first attempt to elucidate breast cancer subtypes by combining microRNA and mRNA expression, immunohistochemical markers, and cancer core pathways. The results can avail the functional studies of the etiology of breast cancer and translated for clinical use given their intrinsic link in terms of immunohistochemistry and survival. This submission consists of microRNA profiles of 115 breast cancer tumors of several subtypes only.

Dataset Information

Breast tumor subtypes correlate with prognosis

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure