**1. Introduction**

164 Breast Cancer – Focusing Tumor Microenvironment, Stem Cells and Metastasis

Wong, Y.C. & Tam, N.N. (2002). Dedifferentiation of stromal smooth muscle as a factor in

Woodworth, C.D.; McMullin, E.; Iglesias, M. & Plowman, G.D. (1995). Interleukin 1 alpha

Wu, Y.;Deng, J.;Rychahou, P.G.;Qiu, S.;Evers, B.M. & Zhou, B.P. (2009). Stabilization of snail

Xouri, G. & Christian, S. (2010). Origin and function of tumor stroma fibroblasts. Seminars in cell & developmental biology, Vol.21, No.1, pp. 40-46, ISSN 1096-3634 Xu, S.-G.; Yan, P.-J. & Shao, Z.-M. (2010). Differential proteomic analysis of a highly

Yaal-Hahoshen, N.; Shina, S.; Leider-Trejo, L.; Barnea, I.; Shabtai, E.L.; Azenshtein, E.;

Yang, L.;Pang, Y. & Moses, H.L. (2010). TGF-beta and immune cells: an important regulatory

Zeisberg, E.M.;Potenta, S.;Xie, L.;Zeisberg, M. & Kalluri, R. (2007). Discovery of endothelial

Zhou, W.; Guo, S. & Gonzalez-Perez, R.R. (2011). Leptin pro-angiogenic signature in breast

Zlotnik, A. (2006). Chemokines and cancer. International journal of cancer, Vol. 119, No. 9,

Zlotnik, A. (2008). New insights on the role of CXCR4 in cancer metastasis. The Journal of

patients. Clinical cancer research, Vol. 12, No. pp. 4474- 4480,

research, Vol.67, No.21, pp. 10123-10128, ISSN 1538-7445

pathology, Vol. 215, No. 3, pp. 211-213, ISSN 1096-9896

Cancer cell, Vol.15, No.5, pp. 416-428, ISSN 1878-3686

4681

No. 7, pp. 2840-2844,

137, 0007-0920

pp. 2026-2029,

No. 10, pp. 1545-1556, 0171-5216

Vol.31, No.6, pp. 220-227, ISSN 1471-4981

prostate carcinogenesis. Differentiation, Vol.70, No.9-10, pp. 633-645, ISSN 0301-

and tumor necrosis factor alpha stimulate autocrine amphiregulin expression and proliferation of human papillomavirus-immortalized and carcinoma-derived cervical epithelial cells. Proceedings of the National Academy of Sciences, Vol. 92,

by NF-kappaB is required for inflammation-induced cell migration and invasion.

metastatic variant of human breast cancer cells using two-dimensional differential gel electrophoresis. Journal of Cancer Research and Clinical Oncology, Vol. 136,

Greenberg, I.; Keydar, I. & Ben-Baruch, A. (2006). The chemokine CCL5 as a potential prognostic factor predicting disease progression in stage II breast cancer

axis in the tumor microenvironment and progression. Trends in immunology,

to mesenchymal transition as a source for carcinoma-associated fibroblasts. Cancer

cancer is linked to IL-1 signalling. British journal of cancer, Vol. 104, No. 1, pp. 128-

Greater than 200,000 new cases of breast cancer cases were diagnosed in 2010 in the United States, with approximately 40,000 women succumbing to the disease (www.cancer.gov). Globally, an estimated 1.38 million new cases of breast cancer were diagnosed in 2008, with greater than 450,000 women succumbing to the disease (Jemal *et al.*, 2011). Despite our improved understanding of breast carcinogenesis, breast cancer remains the second most commonly diagnosed cancer in women behind non-melanoma skin cancer and the second leading cause of death in women behind lung cancer. These epidemiological statistics highlight the overwhelming clinical dilemma of breast cancer and emphasize the need for novel therapeutic targets and prevention strategies. Countless studies in the fields of mammary gland development and breast cancer have led to an appreciation of a breast tumor microenvironment that actively contributes to the heterogeneous nature of breast cancer. The current review will focus on the impact of IL-6 and STAT3 activation in the breast tumor microenvironment and subsequently present rationale for targeting the IL-6/STAT3 signaling pathway in this setting. IL-6 is a quintessential pleiotropic cytokine produced by a diverse number of cell populations, most of which can localize to the breast tumor microenvironment. Excessive IL-6 has been demonstrated in primary breast tumors and breast cancer patient sera and is associated with poor clinical outcomes in breast cancer. These clinical associations are corroborated by emerging preclinical data revealing that IL-6 is a potent growth factor and promotes an epithelial-mesenchymal (EMT) phenotype in breast cancer cells to indicate that IL-6 in the breast tumor microenvironment is clinically relevant. Numerous clinical reports have now demonstrated the safety and efficacy of IL-6 signaling antagonists in multiple diseases, which supports future investigations of these therapies in breast cancer.

Estrogen receptor-alpha (ERα) is a latent cytoplasmic ligand-activated transcription factor utilized by clinicians to subclassify the heterogeneous disease of breast cancer. ERα-positive breast cancer incidence increases up to age 51, the mean age of menopause, and continues to increase until age 80. Conversely, ERα-negative breast cancer incidence plateaus and even slightly decreases at age 51, while demonstrating an increase prior to age 50 comparable to that of ERα-positive disease. This discrepancy between the two incidence rates at menopause produces an inflection in the incidence rate of all breast cancer cases which has been termed Clemmesen's hook (Anderson and Matsuno, 2006). Whereas the prevalence of

Interleukin-6 in the Breast Tumor Microenvironment 167

A normal epithelial tissue can undergo hyperplasia and acquire tumorigenic properties that promote the development of a benign, non-invasive solid tumor known as carcinoma *in situ*. Normal epithelial tissues and non-invasive carcinoma *in situ* tumors are separated from a supportive stromal compartment by an intact basement membrane. Ultimately, carcinoma *in situ* can progress to a malignant, invasive carcinoma, the most common form of human cancer. The panoply of published investigations between the fields of mammary gland development and breast cancer has led to an appreciation for a supportive non-epithelial mammary stroma that mechanically and biologically restrains tumorigenesis. However, tumors of the breast and other epithelial tissues obviously overcome these growth restraints and exploit this stroma to sculpt a vastly divergent tumor stroma. Tumor stroma is generally divided into four main components: tumor vasculature, inflammatory leukocytes, extracellular matrix (ECM) and soluble growth factors, and fibroblasts. Malignant carcinoma cells and tumor stromal cells bi-directionally communicate with one another through paracrine signaling and intercellular contacts in a disorganized ECM to constitute a tumor microenvironment. Tumor-associated fibroblasts (TAF), the predominant stromal cell population within the tumor microenvironment, acquire and sustain an "activated" phenotype that promotes tumor progression (Rasanen and Vaheri, 2010). TAF are capable of enhancing breast tumor growth and metastasis by means of promoting angiogenesis (Orimo *et al.*, 2005), epithelial-mesenchymal transition (EMT) (Martin *et al.*, 2010; Radisky *et al.*, 2005), and progressive genetic instability (Kurose *et al.*, 2001; Moinfar *et al.*, 2000). In contrast, a normal mammary microenvironment can act in a dominant manner to inhibit tumor growth and "revert" the malignant phenotype of breast cancer cells (Kenny and Bissell, 2003). While resident breast tissue fibroblasts can inhabit breast tumors as TAF, breast tumors also recruit distant cell populations that engraft within the breast tumor microenvironment where they actively contribute as TAF. For example, mesenchymal stem cells (MSC), a bone marrow-derived stromal cell population, home to breast cancer cell

Although highly characterized for their protective capacity against infection, inflammatory leukocytes also reside within the tumor microenvironment. In fact, various immune cells are capable of eliminating transformed cells and thus preventing tumorigenesis in a process termed immunosurveillance (Dunn *et al.*, 2004). Whereas acute inflammation may prevent tumorigenesis by promoting an immune response directed against transformed cells, chronic inflammation promotes tumorigenesis. Rudolf Virchow is credited with making the seminal link between chronic inflammation and cancer by noting that human tumor biopsies were often infiltrated with inflammatory cells (Balkwill and Mantovani, 2001). Leukocytes can be detected in non-malignant tumors and carcinomas, including breast cancer (DeNardo and Coussens, 2007), which suggests an ongoing antitumor immune response. Despite the infiltration of leukocytes such as cytotoxic T-cells and NK-cells, the persistence of a tumor demonstrates immune evasion and highlights the local and systemic immune suppressive state of the tumor microenvironment and the tumor-bearing host,

**2. The breast tumor microenvironment** 

xenograft tumors and persist as TAF (Spaeth *et al.*, 2009).

**3. Cancer-associated inflammation** 

respectively.

ERα-positive cells within terminal duct lobular units of the breast of healthy premenopausal women has been reported at 7%, this number is estimated at 42% in postmenopausal women (Shoker *et al.*, 1999). In addition, approximately two-thirds of all breast cancers are diagnosed as ERα-positive, and 75% of postmenopausal breast cancers are ERα-positive (Macedo *et al.*, 2009). Progesterone receptor (PR) and epidermal growth factor receptor 2 (EGFR2; HER2; or ErbB2), a receptor tyrosine kinase involved in cellular proliferation, have also acquired much clinical attention following reports of dismal survival rates in "triple negative" (ERαnegative/PRnegative/HER2 not overexpressed) breast cancer patients. Triple negative breast cancer represents approximately 15 to 20% of all breast cancer cases and can only be treated with standard chemotherapy as it lacks current adjuvant therapeutic targets. Such breast tumors are highly proliferative with a high mitotic index, increased necrosis, elevated apoptosis, and typically are of higher tumor grade. *TP53* gene and p53 protein mutations as well as loss of the Rb tumor suppressor protein are common. Familial breast cancer patients with congenital BRCA1 mutations often present with triple negative breast cancer, as do relatively younger breast cancer patients and African American women. Currently, triple negative breast cancers are associated with a poor prognosis largely due to poor survival rates and early relapse. The fact that these breast tumors respond well if not completely to initial chemotherapy may seem counterintuitive, but enhanced invasiveness, consequent distant metastasis, and residual local recurrence eventually promote poor survival rates (Irvin and Carey, 2008).

Breast cancer most commonly metastasizes to bone, followed by lung, liver, and brain. Perhaps due to the heterogeneity across individual breast cancer cases, few prognostic molecular biomarkers have been demonstrated to accurately predict metastatic potential. One of the most important of these biomarkers is ERα, which is clinically exploited as a predictor of bone metastasis (Kominsky and Davidson, 2006). Whereas ERα-positive breast cancers have a strong tendency to metastasize to bone if at all (James *et al.*, 2003), their ERαnegative counterparts favor visceral organs such as lung and liver (Hess *et al.*, 2003). Primary mammary tumor cell dissemination has been quantified at 3 to 4 x 106 primary tumor cells in circulation per 24 hours per gram of tumor in a rat mammary carcinoma model, which exemplifies the inefficient nature of metastasis (Butler and Gullino, 1975). Although metastasis has been generally accepted as a relatively late event throughout cancer progression, recent work has revealed evidence of early primary tumor cell dissemination, thus refuting this paradigm (Klein, 2009). In particular, it has now been demonstrated that untransformed triple transgenic (doxycycline-inducible K-ras, MYC, and polyoma middle T antigen) mammary epithelial cells are capable of lung colonization when tail vein-injected into immunocompromised female mice on doxycycline. This work showed that untransformed "normal" mammary epithelial cells can colonize ectopic lung tissue, and upon oncogene activation, disseminated mammary epithelial cells within circulation or a foreign host microenvironment are capable of forming tumors at the ectopic site (Podsypanina *et al.*, 2008). Additionally, reports of bone marrow cytokeratin-positive epithelial cells in up to 48% of breast cancer patients without overt metastases also offer support for early primary tumor cell dissemination. Decreased survival in patients with such cells was demonstrated in all studies (Braun *et al.*, 2000; Diel *et al.*, 1996; Gebauer *et al.*, 2001; Pantel *et al.*, 2003; Vannucchi *et al.*, 1998). Furthermore, only 8% of these patients with bone marrow micrometastases exhibited cytokeratin-positive/Ki67-positive cells, suggesting that lack of overt bone metastasis may be due to disseminated tumor cell dormancy (Pantel *et al.*, 2003).

ERα-positive cells within terminal duct lobular units of the breast of healthy premenopausal women has been reported at 7%, this number is estimated at 42% in postmenopausal women (Shoker *et al.*, 1999). In addition, approximately two-thirds of all breast cancers are diagnosed as ERα-positive, and 75% of postmenopausal breast cancers are ERα-positive (Macedo *et al.*, 2009). Progesterone receptor (PR) and epidermal growth factor receptor 2 (EGFR2; HER2; or ErbB2), a receptor tyrosine kinase involved in cellular proliferation, have also acquired much clinical attention following reports of dismal survival rates in "triple

Triple negative breast cancer represents approximately 15 to 20% of all breast cancer cases and can only be treated with standard chemotherapy as it lacks current adjuvant therapeutic targets. Such breast tumors are highly proliferative with a high mitotic index, increased necrosis, elevated apoptosis, and typically are of higher tumor grade. *TP53* gene and p53 protein mutations as well as loss of the Rb tumor suppressor protein are common. Familial breast cancer patients with congenital BRCA1 mutations often present with triple negative breast cancer, as do relatively younger breast cancer patients and African American women. Currently, triple negative breast cancers are associated with a poor prognosis largely due to poor survival rates and early relapse. The fact that these breast tumors respond well if not completely to initial chemotherapy may seem counterintuitive, but enhanced invasiveness, consequent distant metastasis, and residual local recurrence eventually promote poor

Breast cancer most commonly metastasizes to bone, followed by lung, liver, and brain. Perhaps due to the heterogeneity across individual breast cancer cases, few prognostic molecular biomarkers have been demonstrated to accurately predict metastatic potential. One of the most important of these biomarkers is ERα, which is clinically exploited as a predictor of bone metastasis (Kominsky and Davidson, 2006). Whereas ERα-positive breast cancers have a strong tendency to metastasize to bone if at all (James *et al.*, 2003), their ERαnegative counterparts favor visceral organs such as lung and liver (Hess *et al.*, 2003). Primary mammary tumor cell dissemination has been quantified at 3 to 4 x 106 primary tumor cells in circulation per 24 hours per gram of tumor in a rat mammary carcinoma model, which exemplifies the inefficient nature of metastasis (Butler and Gullino, 1975). Although metastasis has been generally accepted as a relatively late event throughout cancer progression, recent work has revealed evidence of early primary tumor cell dissemination, thus refuting this paradigm (Klein, 2009). In particular, it has now been demonstrated that untransformed triple transgenic (doxycycline-inducible K-ras, MYC, and polyoma middle T antigen) mammary epithelial cells are capable of lung colonization when tail vein-injected into immunocompromised female mice on doxycycline. This work showed that untransformed "normal" mammary epithelial cells can colonize ectopic lung tissue, and upon oncogene activation, disseminated mammary epithelial cells within circulation or a foreign host microenvironment are capable of forming tumors at the ectopic site (Podsypanina *et al.*, 2008). Additionally, reports of bone marrow cytokeratin-positive epithelial cells in up to 48% of breast cancer patients without overt metastases also offer support for early primary tumor cell dissemination. Decreased survival in patients with such cells was demonstrated in all studies (Braun *et al.*, 2000; Diel *et al.*, 1996; Gebauer *et al.*, 2001; Pantel *et al.*, 2003; Vannucchi *et al.*, 1998). Furthermore, only 8% of these patients with bone marrow micrometastases exhibited cytokeratin-positive/Ki67-positive cells, suggesting that lack of overt bone metastasis may be due to disseminated tumor cell

negative/HER2 not overexpressed) breast cancer patients.

negative" (ERα-

negative/PR-

survival rates (Irvin and Carey, 2008).

dormancy (Pantel *et al.*, 2003).
