Integrative clustering reveals a novel split in the luminal A subtype of breast cancer with impact on outcome.

BACKGROUND: Breast cancer is a heterogeneous disease at the clinical and molecular level. In this study we integrate classifications extracted from five different molecular levels in order to identify integrated subtypes. METHODS: Tumor tissue from 425 patients with primary breast cancer from the Oslo2 study was cut and blended, and divided into fractions for DNA, RNA and protein isolation and metabolomics, allowing the acquisition of representative and comparable molecular data. Patients were stratified into groups based on their tumor characteristics from five different molecular levels, using various clustering methods. Finally, all previously identified and newly determined subgroups were combined in a multilevel classification using a "cluster-of-clusters" approach with consensus clustering. RESULTS: Based on DNA copy number data, tumors were categorized into three groups according to the complex arm aberration index. mRNA expression profiles divided tumors into five molecular subgroups according to PAM50 subtyping, and clustering based on microRNA expression revealed four subgroups. Reverse-phase protein array data divided tumors into five subgroups. Hierarchical clustering of tumor metabolic profiles revealed three clusters. Combining DNA copy number and mRNA expression classified tumors into seven clusters based on pathway activity levels, and tumors were classified into ten subtypes using integrative clustering. The final consensus clustering that incorporated all aforementioned subtypes revealed six major groups. Five corresponded well with the mRNA subtypes, while a sixth group resulted from a split of the luminal A subtype; these tumors belonged to distinct microRNA clusters. Gain-of-function studies using MCF-7 cells showed that microRNAs differentially expressed between the luminal A clusters were important for cancer cell survival. These microRNAs were used to validate the split in luminal A tumors in four independent breast cancer cohorts. In two cohorts the microRNAs divided tumors into subgroups with significantly different outcomes, and in another a trend was observed. CONCLUSIONS: The six integrated subtypes identified confirm the heterogeneity of breast cancer and show that finer subdivisions of subtypes are evident. Increasing knowledge of the heterogeneity of the luminal A subtype may add pivotal information to guide therapeutic choices, evidently bringing us closer to improved treatment for this largest subgroup of breast cancer.

Tumor expression, plasma levels and genetic polymorphisms of the coagulation inhibitor TFPI are associated with clinicopathological parameters and survival in breast cancer, in contrast to the coagulation initiator TF.

INTRODUCTION: Hypercoagulability in malignancy increases the risk of thrombosis, but is also involved in cancer progression. Experimental studies suggest that tissue factor (TF) and tissue factor pathway inhibitor (TFPI) are involved in cancer biology as a tumor- promoter and suppressor, respectively, but the clinical significance is less clear. Here, we aimed to investigate the clinical relevance of TF and TFPI genetic and phenotypic diversity in breast cancer. METHODS: The relationship between tumor messenger RNA (mRNA) expression and plasma levels of TF and TFPI (α and ß), tagging single nucleotide polymorphisms (tagSNPs) in F3 (TF) (n=6) and TFPI (n=18), and clinicopathological characteristics and molecular tumor subtypes were explored in 152 treatment naive breast cancer patients. The effect of tumor expressed TF and TFPIα and TFPIß on survival was investigated in a merged breast cancer dataset of 1881 patients. RESULTS: Progesterone receptor negative patients had higher mRNA expression of total TFPI (α+ß) (P=0.021) and TFPIß (P=0.014) in tumors. TF mRNA expression was decreased in grade 3 tumors (P=0.003). In plasma, total TFPI levels were decreased in patients with larger tumors (P=0.013). SNP haplotypes of TFPI, but not TF, were associated with specific clinicopathological characteristics like tumor size (odds ratio (OR) 3.14, P=0.004), triple negativity (OR 2.4, P=0.004), lymph node spread (OR 3.34, P=0.006), and basal-like (OR 2.3, P=0.011) and luminal B (OR 3.5, P=0.005) molecular tumor subtypes. Increased expression levels of TFPIα and TFPIß in breast tumors were associated with better outcome in all tumor subtypes combined (P=0.007 and P=0.005) and in multiple subgroups, including lymph node positive subjects (P=0.006 and P=0.034). CONCLUSIONS: This study indicates that genetic and phenotypic variation of both TFPIα and TFPIß, more than TF, are markers of cancer progression. Together with the previously demonstrated tumor suppressor effects of TFPI, the beneficial effect of tumor expressed TFPI on survival, renders TFPI as a potential anticancer agent, and the clinical significance of TFPI in cancer deserves further investigation.

Evaluating immune response in vitro in a relevant microenvironment: a high-throughput microfluidic model for clinical screening.

Aim: Functional screening of new pharmaceutical compounds requires clinically relevant models to monitor essential cellular and immune responses during cancer progression, with or without treatment. Beyond survival, the emergence of resistant tumor cell clones should also be considered, including specific properties related to plasticity, such as invasiveness, stemness, escape from programmed cell death, and immune response. Numerous pathways are involved in these processes. Defining the relevant ones in the context of a specific tumor type will be key to designing an appropriate combination of inhibitors. However, the diversity and potential redundancy of these pathways remain a challenge for therapy. Methods: A new microfluidic device developed by Okomera was dedicated to the screening of drug treatment for breast cancer. This microchip includes 150 droplet-trapping microwells, offering multi-chip settings and multiple treatment choices. Results: After validating the system with established cell lines and a panel of drugs used clinically at Gustave Roussy, preclinical experiments were initiated including patient-derived xenograft (PDX) and primary tumor cells-derived tumoroids with the collaboration of Gustave Roussy clinicians. Tumor-isolated lymphocytes were also added to the tumoroids, using secondary droplets in proof-of-concept experiments. Conclusions: These results show the relevance of the methodology for screening large numbers of drugs, a wide range of doses, and multiple drug combinations. This methodology will be used for two purposes: 1) new drug screening from the compound library, using the high throughput potential of the chip; and 2) pre-clinical assay for a two-weeks response for personalized medicine, allowing evaluation of drug combinations to flag an optimized treatment with potential clinical application.

Drug-screening and genomic analyses of HER2-positive breast cancer cell lines reveal predictors for treatment response.

BACKGROUND: Approximately 15%-20% of all diagnosed breast cancers are characterized by amplified and overexpressed HER2 (= ErbB2). These breast cancers are aggressive and have a poor prognosis. Although improvements in treatment have been achieved after the introduction of trastuzumab and lapatinib, many patients do not benefit from these drugs. Therefore, in-depth understanding of the mechanisms behind the treatment responses is essential to find alternative therapeutic strategies. MATERIALS AND METHODS: Thirteen HER2 positive breast cancer cell lines were screened with 22 commercially available compounds, mainly targeting proteins in the ErbB2-signaling pathway, and molecular mechanisms related to treatment sensitivity were sought. Cell viability was measured, and treatment responses between the cell lines were compared. To search for response predictors and genomic and transcriptomic profiling, PIK3CA mutations and PTEN status were explored and molecular features associated with drug sensitivity sought. RESULTS: The cell lines were divided into three groups according to the growth-retarding effect induced by trastuzumab and lapatinib. Interestingly, two cell lines insensitive to trastuzumab (KPL4 and SUM190PT) showed sensitivity to an Akt1/2 kinase inhibitor. These cell lines had mutation in PIK3CA and loss of PTEN, suggesting an activated and druggable Akt-signaling pathway. Expression levels of five genes (CDC42, MAPK8, PLCG1, PTK6, and PAK6) were suggested as predictors for the Akt1/2 kinase-inhibitor response. CONCLUSION: Targeting the Akt-signaling pathway shows promise in cell lines that do not respond to trastuzumab. In addition, our results indicate that several molecular features determine the growth-retarding effects induced by the drugs, suggesting that parameters other than HER2 amplification/expression should be included as markers for therapy decisions.

Metabolic clusters of breast cancer in relation to gene- and protein expression subtypes.

BACKGROUND: The heterogeneous biology of breast cancer leads to high diversity in prognosis and response to treatment, even for patients with similar clinical diagnosis, histology, and stage of disease. Identifying mechanisms contributing to this heterogeneity may reveal new cancer targets or clinically relevant subgroups for treatment stratification. In this study, we have merged metabolite, protein, and gene expression data from breast cancer patients to examine the heterogeneity at a molecular level. METHODS: The study included primary tumor samples from 228 non-treated breast cancer patients. High-resolution magic-angle spinning magnetic resonance spectroscopy (HR MAS MRS) was performed to extract the tumors metabolic profiles further used for hierarchical cluster analysis resulting in three significantly different metabolic clusters (Mc1, Mc2, and Mc3). The clusters were further combined with gene and protein expression data. RESULTS: Our result revealed distinct differences in the metabolic profile of the three metabolic clusters. Among the most interesting differences, Mc1 had the highest levels of glycerophosphocholine (GPC) and phosphocholine (PCho), Mc2 had the highest levels of glucose, and Mc3 had the highest levels of lactate and alanine. Integrated pathway analysis of metabolite and gene expression data uncovered differences in glycolysis/gluconeogenesis and glycerophospholipid metabolism between the clusters. All three clusters had significant differences in the distribution of protein subtypes classified by the expression of breast cancer-related proteins. Genes related to collagens and extracellular matrix were downregulated in Mc1 and consequently upregulated in Mc2 and Mc3, underpinning the differences in protein subtypes within the metabolic clusters. Genetic subtypes were evenly distributed among the three metabolic clusters and could therefore contribute to additional explanation of breast cancer heterogeneity. CONCLUSIONS: Three naturally occurring metabolic clusters of breast cancer were detected among primary tumors from non-treated breast cancer patients. The clusters expressed differences in breast cancer-related protein as well as genes related to extracellular matrix and metabolic pathways known to be aberrant in cancer. Analyses of metabolic activity combined with gene and protein expression provide new information about the heterogeneity of breast tumors and, importantly, the metabolic differences infer that the clusters may be susceptible to different metabolically targeted drugs.

Integrated analysis reveals microRNA networks coordinately expressed with key proteins in breast cancer.

BACKGROUND: The role played by microRNAs in the deregulation of protein expression in breast cancer is only partly understood. To gain insight, the combined effect of microRNA and mRNA expression on protein expression was investigated in three independent data sets. METHODS: Protein expression was modeled as a multilinear function of powers of mRNA and microRNA expression. The model was first applied to mRNA and protein expression for 105 selected cancer-associated genes and to genome-wide microRNA expression from 283 breast tumors. The model considered both the effect of one microRNA at a time and all microRNAs combined. In the latter case the Lasso penalized regression method was applied to detect the simultaneous effect of multiple microRNAs. RESULTS: An interactome map for breast cancer representing all direct and indirect associations between the expression of microRNAs and proteins was derived. A pattern of extensive coordination between microRNA and protein expression in breast cancer emerges, with multiple clusters of microRNAs being associated with multiple clusters of proteins. Results were subsequently validated in two independent breast cancer data sets. A number of the microRNA-protein associations were functionally validated in a breast cancer cell line. CONCLUSIONS: A comprehensive map is derived for the co-expression in breast cancer of microRNAs and 105 proteins with known roles in cancer, after filtering out the in-cis effect of mRNA expression. The analysis suggests that group action by several microRNAs to deregulate the expression of proteins is a common modus operandi in breast cancer.

High-throughput 3D screening reveals differences in drug sensitivities between culture models of JIMT1 breast cancer cells.

The traditional method for studying cancer in vitro is to grow immortalized cancer cells in two-dimensional monolayers on plastic. However, many cellular features are impaired in these artificial conditions, and large changes in gene expression compared to tumors have been reported. Three-dimensional cell culture models have become increasingly popular and are suggested to be better models than two-dimensional monolayers due to improved cell-to-cell contact and structures that resemble in vivo architecture. The aim of this study was to develop a simple high-throughput three-dimensional drug screening method and to compare drug responses in JIMT1 breast cancer cells when grown in two dimensions, in poly(2-hydroxyethyl methacrylate) induced anchorage-independent three-dimensional models, and in Matrigel three-dimensional cell culture models. We screened 102 compounds with multiple concentrations and biological replicates for their effects on cell proliferation. The cells were either treated immediately upon plating, or they were allowed to grow in three-dimensional cultures for 4 days before the drug treatment. Large variations in drug responses were observed between the models indicating that comparisons of culture model-influenced drug sensitivities cannot be made based on the effects of a single drug. However, we show with the 63 most prominent drugs that, in general, JIMT1 cells grown on Matrigel were significantly more sensitive to drugs than cells grown in two-dimensional cultures, while the responses of cells grown in poly(2-hydroxyethyl methacrylate) resembled those of the two-dimensional cultures. Furthermore, comparing the gene expression profiles of the cell culture models to xenograft tumors indicated that cells cultured in Matrigel and as xenografts most closely resembled each other. In this study, we also suggest that three-dimensional cultures can provide a platform for systematic experimentation of larger compound collections in a high-throughput mode and be used as alternatives to traditional two-dimensional screens for better comparability to the in vivo state.