Tumor cells are commonly spread to tumor-draining lymph nodes in many types of cancer. There is ample evidence that lymphangiogenesis is actively induced by tumor cells and lymphangiogenic growth factors play an important role in lymphangiogenesis and distal organ metastasis [
15‐
17]. For example, the proteolytic matured VEGF-C and VEGF-D interact with VEGFR-3 and upregulate lymphangiogenic activity [
18,
19]; the VEGF-A-VEGFR-2 axis also induces tumor lymphangiogenesis by inducing LEC proliferation [
20,
21]. Angiogenin 1 (Ang1) and Ang2 were also shown to be important for stimulating lymphangiogenesis and lymphatic metastasis in a tumor xenograft model [
22,
23]. Hepatocyte growth factor receptor (HGF c-MET) is also known to be an inducer of lymphangiogenesis in vivo [
24‐
26]. Studies by our laboratory and others showed that fibroblast growth factor (FGF), epidermal growth factor (EGF), PDGF, and insulin-like growth factor (IGF) play critical roles in lymphangiogenesis [
13,
27‐
30]. In our previous studies, we have shown that crosstalk between MDA-MB-231-LN cells and LECs induce upregulation of EGF to promote MDA-MB-231-LN cell proliferation and LECs co-injected with MDA-MB-231-LN cells into nude mice enhanced tumor growth. In addition, we demonstrated that an interaction of MDA-MB-231-LN and LEC stimulated PDGF-BB expression and recruited pericytes to the neovasculature in a xenograft mouse model; SU16f, a PDGFRβ inhibitor, inhibited the recruitment of pericytes in this model [
13]. It has been reported that the CCL5-CCR5 axis enhanced metastasis of basal breast cancer cells [
31‐
33], but detailed mechanisms of how this happens are not yet clear. We have tried to clarify these mechanisms in this and earlier studies. In a previous study, we discovered that IL-6 secreted by TNBC cells upregulates CCL5 and VEGF in LECs through the IL-6-STAT3 signaling pathway. Additionally, we showed that the secreted CCL5 recruited CCR5-positive TNBC breast cancer to LNs resulting in LN angiogenesis and lung metastasis. In addition, the increased levels of VEGF induced LN angiogenesis and supported tumor cell extravasation into the lung. In agreement with these results, we have also found a significant inhibitory effect on TNBC metastasis by the following manipulations: depleting IL-6 in the tumor-conditioned medium injected into mice prior to tumor inoculation using an anti-IL-6 antibody, inhibiting CCL5 by maraviroc, inhibiting VEGF by an anti-VEGF antibody, and inhibiting STAT3 by S3I-201 [
8]. Maraviroc exerts its anti-retroviral activity by inhibiting the CCR5 receptor and it is commonly utilized as a treatment for HIV infection. Our results show that maraviroc inhibits TNBC metastasis so we propose repurposing this drug to treat patients with TNBC. In order to produce even more effective treatments for TNBC using repurposed drugs, we tested tocilizumab, a Food and Drug Administration (FDA)-approved anti-inflammatory IL-6 receptor inhibitor drug for the treatment of rheumatoid arthritis. Application of these two drugs allowed us to directly test the hypothesis that drugs against both IL-6 and CCL5 could be used to treat TNBC. We discovered that co-cultures of MDA-MB-231-LN cells with LECs enhanced proliferation of MDA-MB-231-LN cells. It has been reported that autocrine IL-6 is a key promoter of TNBC cell proliferation [
34,
35]. In support of our hypothesis, in our experiments, tocilizumab significantly decreased the viability of TNBC cells and the combination of tocilizumab and maraviroc decreased cellular viability even more (Fig.
1). Consistent with this result, we observed no significant difference in the viability of TNBC cells in the presence of maraviroc. Our data demonstrate that IL-6 signaling in the crosstalk between TNBC cells and LECs is a major determinant of TNBC cell proliferation and viability.
We next investigated the effect of maraviroc and tocilizumab on the migration of TNBC cells induced by (TCM-LEC)CM. The results indicate that the migration of TNBC cells is dependent on CCL5-CCR5 signaling and that IL-6 indirectly contributes to migration of TNBC cells by regulating CCL5 expression in LECs (Fig.
2). We tested agents targeting the IL-6 and CCR5 pathways in athymic mice with TNBC tumor xenografts to extend these findings. The growth of MDA-MB-231-LN tumors was inhibited by cMR16-1, whereas maraviroc had no effect on tumor growth, consistent with our observation that maraviroc had no effect on the proliferation of MDA-MB-231-LN cells. The combination of maraviroc and cMR16-1 resulted in greater inhibition of tumor growth compared to the single agents (Fig.
3). Next, we were interested in determining whether maraviroc and cMR16-1 could inhibit TNBC tumor metastasis. Thoracic metastasis was detected in all control group mice while mice treated with either maraviroc or cMR16-1 had low incidences of metastasis. Strikingly, we observed no metastasis in any of the mice treated with a combination of maraviroc and cMR16-1 (Fig.
4). These observations demonstrate that CCL5-CCR5 and IL-6-IL-6R signaling between TNBC cells and LECs plays a critical role in TNBC tumor growth and metastasis.