Tina Gruosso

Identification of interacting stromal axes in triple-negative breast cancer

Triple-negative breast cancer (TNBC) is a molecularly heterogeneous cancer that is difficult to treat. Despite the role it may play in tumor progression and response to therapy, microenvironmental (stromal) heterogeneity in TNBC has not been well characterized. To address this challenge, we investigated the transcriptome of tumor-associated stroma isolated from TNBC (n = 57). We identified four stromal axes enriched for T cells (T), B cells (B), epithelial markers (E), or desmoplasia (D). Our analysis method (STROMA4) assigns a score along each stromal axis for each patient and then combined the axis scores to subtype patients. Analysis of these subtypes revealed that prognostic capacity of the B, T, and E scores was governed by the D score. When compared with a previously published TNBC subtyping scheme, the STROMA4 method better captured tumor heterogeneity and predicted patient benefit from therapy with increased sensitivity. This approach produces a simple ontology that captures TNBC heterogeneity and informs how tumor-associated properties interact to affect prognosis. Cancer Res; 77(17); 4673-83. ©2017 AACR.

Abstract A69: The immunological environment in triple negative breast cancer; impact on clinical outcomes towards a prognostic clinical test

It is now accepted that changes in the normal cells that constitute the tumor microenvironment (TME) play important roles in determining cancer progression and ultimate outcome. We and others have established that an immune signature, and more recently that the ability of CD8+ T cells to infiltrate the tumor bed, are correlated with good outcome in triple negative breast cancer. Importantly, since the cytotoxic effector functions of CD8+ T cells are mainly contact-dependent, the tumor microenvironment may act to inhibit access to the epithelial tumor bed, leading to uncontrolled tumor growth. Therefore, in order to determine the role that the TME plays on mediating the entry of CD8+ T cells into the epithelial tumor bed, our group performed laser-capture microdissection to separate the tumor epithelial compartment from the surrounding tumor-associated stroma from the primary tumor of 56 triple negative breast cancer patients and followed by gene expression profiling. Our aims are to provide clinical validation of predicted CD8+ T cell localisation based on our bioinformatic analysis, test our predictions that the identified stroma signature correlates with CD8+ T cell retention in other cohort of TN breast cancer patients and develop suitable markers of this stroma, and develop a muliplex based assay integrating CD8+ T cell localisation with stromal markers suitable for a retrospective study to validate predictions with outcome. Using gene expression profiling, our data have identified a canonical gene expression signature which correlates with retention of CD8+ T cells in the stroma and poor outcome. Thus, our results demonstrate that gene expression analysis of clinical triple negative breast cancer samples can predict the location of CD8+ T cells within the tumor and in turn the outcome. Citation Format: Mathieu Gigoux, Tina Gruosso, Nicholas Bertos, Sadiq Saleh, Atilla Omeroglu, Hong Zhao, Margarita Souleimanova, Sarkis Meterissian, Michael Hallett, Morag Park. The immunological environment in triple negative breast cancer; impact on clinical outcomes towards a prognostic clinical test. [abstract]. In: Proceedings of the AACR Special Conference: Tumor Immunology and Immunotherapy: A New Chapter; December 1-4, 2014; Orlando, FL. Philadelphia (PA): AACR; Cancer Immunol Res 2015;3(10 Suppl):Abstract nr A69.

Infiltration of CD8+ T cells into tumor-cell clusters in Triple Negative Breast Cancer

Infiltration of CD8+ T lymphocytes into solid tumors is associated with good prognosis in various types of cancer, including Triple Negative Breast Cancers (TNBC). However, the mechanisms underlying different infiltration-levels are largely unknown. Here, we have characterized the spatial profile of CD8+ T cells around tumorcell clusters in the core and margin regions in TNBC. Combining mathematical modeling and data analysis, we propose that there exists a possible chemo-repellent inside tumor-cell clusters, which prevents CD8+ T cells from infiltrating into tumor-cell clusters. Furthermore, investigation into the properties of collagen fibers suggests that variations in desmoplastic elements does not limit infiltration of CD8+ T lymphocytes into tumor-cell clusters. This is consistent with the prediction of our mathematical modeling analysis whereby CD8+ T cells are predicted to infiltrate the fibrotic barrier and their infiltration into tumor clusters is governed by other mechanisms involving a local repellent.

KIBRA (WWC1) Is a Metastasis Suppressor Gene Affected by Chromosome 5q Loss in Triple-Negative Breast Cancer

Summary Triple-negative breast cancers (TNBCs) display a complex spectrum of mutations and chromosomal aberrations. Chromosome 5q (5q) loss is detected in up to 70% of TNBCs, but little is known regarding the genetic drivers associated with this event. Here, we show somatic deletion of a region syntenic with human 5q33.2–35.3 in a mouse model of TNBC. Mechanistically, we identify KIBRA as a major factor contributing to the effects of 5q loss on tumor growth and metastatic progression. Re-expression of KIBRA impairs metastasis in vivo and inhibits tumorsphere formation by TNBC cells in vitro. KIBRA functions co-operatively with the protein tyrosine phosphatase PTPN14 to trigger mechanotransduction-regulated signals that inhibit the nuclear localization of oncogenic transcriptional co-activators YAP/TAZ. Our results argue that the selective advantage produced by 5q loss involves reduced dosage of KIBRA, promoting oncogenic functioning of YAP/TAZ in TNBC.

CrosstalkNet: A Visualization Tool for Differential Co-expression Networks and Communities

Variations in physiological conditions can rewire molecular interactions between biological compartments, which can yield novel insights into gain or loss of interactions specific to perturbations of interest. Networks are a promising tool to elucidate intercellular interactions, yet exploration of these large-scale networks remains a challenge due to their high dimensionality. To retrieve and mine interactions, we developed CrosstalkNet, a user friendly, web-based network visualization tool that provides a statistical framework to infer condition-specific interactions coupled with a community detection algorithm for bipartite graphs to identify significantly dense subnetworks. As a case study, we used CrosstalkNet to mine a set of 54 and 22 gene-expression profiles from breast tumor and normal samples, respectively, with epithelial and stromal compartments extracted via laser microdissection. We show how CrosstalkNet can be used to explore large-scale co-expression networks and to obtain insights into the biological processes that govern cross-talk between different tumor compartments.Significance: This web application enables researchers to mine complex networks and to decipher novel biological processes in tumor epithelial-stroma cross-talk as well as in other studies of intercompartmental interactions. Cancer Res; 78(8); 2140-3. ©2018 AACR.

CrosstalkNet: mining large-scale bipartite co-expression networks to characterize epi-stroma crosstalk

Background: Over the last several years, we have witnessed the metamorphosis of network biology from being a mere representation of molecular interactions to models enabling inference of complex biological processes. Networks provide promising tools to elucidate intercellular interactions that contribute to the functioning of key biological pathways in a cell. However, the exploration of these large-scale networks remains a challenge due to their high-dimensionality. Results: CrosstalkNet is a user friendly, web-based network visualization tool to retrieve and mine interactions in large-scale bipartite co-expression networks. In this study, we discuss the use of gene co-expression networks to explore the rewiring of interactions between tumor epithelial and stromal cells. We show how CrosstalkNet can be used to efficiently visualize, mine, and interpret large co-expression networks representing the crosstalk occurring between the tumour and its microenvironment. Conclusion: CrosstalkNet serves as a tool to assist biologists and clinicians in exploring complex, large interaction graphs to obtain insights into the biological processes that govern the tumor epithelial-stromal crosstalk. A comprehensive tutorial along with case studies are provided with the application. Availability: The web-based application is available at the following location: http://epistroma.pmgenomics.ca/app/. The code is open-source and freely available from http://github.com/bhklab/EpiStroma-webapp. Contact: bhaibeka@uhnresearch.ca

Abstract A15: Mechanisms of CD8+ T cell immunosuppression in triple negative breast cancer

Triple negative breast cancer (TNBC), defined as tumors lacking expression of the estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor 2 (HER2), are especially difficult to treat effectively. While ER+ and HER2+ breast cancer subtypes can be treated with Tamoxifen and Herceptin, respectively, there are no targeted therapies for TNBC patients. Furthermore, while only 20-30% of TNBC patients respond to chemotherapy in the neoadjuvant setting, overall outcome remains poor for non-responding patients. However, mounting evidence suggests that immune-checkpoint inhibitor immunotherapies may be especially promising for TNBC patients. We and others have shown that the presence of CD8+ T cells, a crucial component of the cytotoxic arm of the adaptive immune response, is a sign of good clinical outcome in TNBC patients. However, good outcome only correlates with CD8+ T cell invasion of the tumor parenchyma. Here we show that some patients have an accumulation of CD8+ T cells in the surrounding tumor-associated stroma, but not the tumor epithelium, and these patients responded as poorly as patients with no CD8+ T cells at all. Yet how cancer associated fibroblasts (CAFs), a dominant cell type of the tumor-associated stroma, affects CD8+ T cell invasion into the tumor epithelium is still poorly understood. To identify potential stroma-dependent mechanisms that potentiate or inhibit CD8+ T cells invasion into the tumor epithelium, we performed gene expression profiling of laser-capture microdissected tumor-associated stroma (and matched epithelium) from 56 TNBC cases. Here we identify several key stromal features that may explain the accumulation of CD8+ T cells outside of the tumor epithelium. Preliminary data by immunohistochemistry and immunofluorescence validate some key stromal features and decipher the implication of other immune cell types in CD8+ T cells lack of tumor epithelium infiltration. These key stromal features that impair CD8+ T cell infiltration into the tumor in some patients might explain the relative low efficiency of immunotherapies in TNBC patients (20% of patients respond). One could speculate that targeting these key stromal features would allow a significant CD8+ T cell infiltration into the tumor and thus sensitize patients to immunotherapies. Citation Format: Tina Gruosso, Mathieu Gigoux, Nicholas Bertos, Sadiq Saleh, Atilla Omeroglu, Dongmei Zuo, Hong Zhao, Margarita Souleimanova, Valerie Weaver, Sarkis Meterissian, Michael Hallett, Morag Park. Mechanisms of CD8+ T cell immunosuppression in triple negative breast cancer. [abstract]. In: Proceedings of the AACR Special Conference: Function of Tumor Microenvironment in Cancer Progression; 2016 Jan 7–10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2016;76(15 Suppl):Abstract nr A15.

Abstract 720: Novel prognostic stromal subtypes in triple-negative breast cancer

Breast cancer is a heterogeneous disease in terms of presentation, morphology, molecular profile and response to therapy. Gene expression profiling has identified intrinsic molecular subtypes that are associated with clinical markers (ER, PR, HER2) as well as prognosis and survival. However, it is well established that the intrinsic molecular profiles of breast tumors are not sufficient to perfectly predict disease outcome. Increasing evidence indicates that characteristics of the breast stroma influence tumor progression and response to therapy. Previous work in our lab has demonstrated that gene expression signatures in human stroma can predict outcome of breast cancer patients independently of clinical parameters and molecular subtypes. In this study, we expand our findings by focusing on a previously underrepresented subset of breast tumors that have no detectable ER, PR or HER2 (termed Triple-Negative, TN). TN tumors, which represent approximately 15% of all breast cancers, are typically associated with poor outcome. However, the contribution of the stroma to the underlying heterogeneity of TN breast cancer and its corresponding influence on therapeutic response is not well understood. To address this, we isolated TN tumor epithelial and stromal tissues by laser capture microdissection and subjected them to gene expression profiling. Class discovery revealed distinct gene-clusters (stromal properties) which are associated with prognosis in TNBC whole tumor samples. Analysis of the genes comprising each stromal property suggests that the properties primarily represent the prevalence of distinct cell types, namely T cells, B cells, activated fibroblasts, and myoepithelial cells. Importantly, these properties are not mutually exclusive, i.e. some tumors are associated with multiple stromal properties. While confirming the heterogeneity of TN-associated stroma, this also indicates that a multi-parameter classification better reflects the true nature of the tumor microenvironment. This project provides the first integrated in-depth analysis of the contribution of tumor stromal processes to TN disease heterogeneity, and positions the tumor microenvironment for therapeutic intervention. Citation Format: Crista Thompson, Sadiq M. Saleh, Nicholas Bertos, Mathieu Gigoux, Tina Gruosso, Margarita Souleimanova, Hong Zhao, Michael T. Hallett, Morag Park. Novel prognostic stromal subtypes in triple-negative breast cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 720.

Abstract IA23: Deconvolution of the triple-negative breast cancer microenvironment

Breast cancer heterogeneity is one of the principal obstacles both to predicting outcome and to determining an effective course of treatment for this disease. Although genomic technologies have been used to gain a better understanding, by identifying gene expression signatures associated with clinical outcome and breast cancer subtypes, relatively little is known about heterogeneity in the tumor microenvironment. It is now accepted that changes in the normal cells that constitute the tumor microenvironment (TME) play important roles in determining cancer progression and ultimate outcome. We and others have established that an immune gene expression signature correlates with good outcome in triple negative breast cancer (TNBC), yet fails to accurately predict outcome in all patients. Examining stromal heterogeneity in TNBC has identified four distinct stromal clusters, three of which are prognostic, contain distinct immune signatures and a signature of fibrosis. Lymphocytic infiltration and access to tumor parenchyma is not well understood due to high levels of spatial heterogeneity within tumors. We show that location of CD8+T cells is strongly influenced by TME subtypes and identify gene expression signatures predictive of distinct CD8+T cell localization and patient outcome. Since mounting evidence suggests that immune-checkpoint inhibitor immunotherapies may be promising for only a subset of TNBC patients, highlights the importance of understanding how the TME influences CD8+T cell location. Citation Format: Sadiq Saleh, Tina Gruosso, Mathieu Gigoux, Nicholas Bertos, Atilla Omeroglu, Dongmei Zuo, Sarkis Meterissian, Michael Hallett, Morag Park. Deconvolution of the triple-negative breast cancer microenvironment. [abstract]. In: Proceedings of the AACR Special Conference: Function of Tumor Microenvironment in Cancer Progression; 2016 Jan 7–10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2016;76(15 Suppl):Abstract nr IA23.

The Human Adenovirus Type 5 E4orf4 Protein Targets Two Phosphatase Regulators of the Hippo Signaling Pathway

ABSTRACT When expressed alone at high levels, the human adenovirus E4orf4 protein exhibits tumor cell-specific p53-independent toxicity. A major E4orf4 target is the B55 class of PP2A regulatory subunits, and we have shown recently that binding of E4orf4 inhibits PP2AB55 phosphatase activity in a dose-dependent fashion by preventing access of substrates (M. Z. Mui et al., PLoS Pathog 9:e1003742, 2013, http://dx.doi.org/10.1371/journal.ppat.1003742). While interaction with B55 subunits is essential for toxicity, E4orf4 mutants exist that, despite binding B55 at high levels, are defective in cell killing, suggesting that other essential targets exist. In an attempt to identify additional targets, we undertook a proteomics approach to characterize E4orf4-interacting proteins. Our findings indicated that, in addition to PP2AB55 subunits, ASPP-PP1 complex subunits were found among the major E4orf4-binding species. Both the PP2A and ASPP-PP1 phosphatases are known to positively regulate effectors of the Hippo signaling pathway, which controls the expression of cell growth/survival genes by dephosphorylating the YAP transcriptional coactivator. We find here that expression of E4orf4 results in hyperphosphorylation of YAP, suggesting that Hippo signaling is affected by E4orf4 interactions with PP2AB55 and/or ASPP-PP1 phosphatases. Furthermore, knockdown of YAP1 expression was seen to enhance E4orf4 killing, again consistent with a link between E4orf4 toxicity and inhibition of the Hippo pathway. This effect may in fact contribute to the cancer cell specificity of E4orf4 toxicity, as many human cancer cells rely heavily on the Hippo pathway for their enhanced proliferation. IMPORTANCE The human adenovirus E4orf4 protein has been known for some time to induce tumor cell-specific death when expressed at high levels; thus, knowledge of its mode of action could be of importance for development of new cancer therapies. Although the B55 form of the phosphatase PP2A has long been known as an essential E4orf4 target, genetic analyses indicated that others must exist. To identify additional E4orf4 targets, we performed, for the first time, a large-scale affinity purification/mass spectrometry analysis of E4orf4 binding partners. Several additional candidates were detected, including key regulators of the Hippo signaling pathway, which enhances cell viability in many cancers, and results of preliminary studies suggested a link between inhibition of Hippo signaling and E4orf4 toxicity.