Tumor-Immune interactions: implications for cancer therapy.

The tumor microenvironment is often compared to a complex organ, consisting of many cell types including immune cells, activated fibroblasts and endothelial cells in addition to the cancer cells. Thus, tumor progression is not solely determined by changes within the cancer cells, but also by their interactions with the surrounding stroma. In particular, interactions with the immune system are increasingly being recognized to play a central role in tumor progression. Furthermore, heterogeneity among the cancer cells also determines how and when tumors progress. Tumor initiating cells (TICs) are thought to be responsible for many features of cancer, including resistance to chemotherapy, recurrence following initial response, cellular heterogeneity and metastasis. Given the importance of immune cells and TICs, it comes as no surprise that cross-talks between these cell types could determine the course of disease. Understanding these interactions will be essential to improving therapeutic development.Tumor-associated macrophages (TAMs), particularly of the M2 type, have long been associated with poor prognosis in various cancers. However, until recently, the precise subset of cancer cells interacting with macrophages has been only partially characterized. Several recent reports show that a number of human cancer cell lines can “educate” macrophages to stimulate TIC survival and tumorigenesis via mechanisms involving STAT3 and IL-6.1 Using a spontaneous murine model of uveal melanoma, we have shown that M2 macrophages stimulate CD34− but not CD34+ melanoma TICs, illustrating the heterogeneous response of cancer cells to micro-environmental factors.2 We showed that TGFβ and polyamines produced by TAMs stimulate TIC survival and proliferation. Importantly, the M2 phenotype of the macrophages, i.e. production of TGFβ and polyamines, is sufficient for TIC stimulation and prior exposure to cancer cells may not be necessary. This may explain why certain organs, such as the lungs and liver, that contain a high density of M2-like macrophages and high levels of TGFβ and polyamines in physiological conditions, are more susceptible to metastasis: circulating tumor cells reaching these organs may have more chance to survive and initiate tumor growth.We also demonstrated that the phenotype of melanoma TICs is highly flexible. A short-term in vivo depletion of macrophages with treatments against the colony stimulatory factor-1 receptor (CSF-1R) eliminates TAM-responsive TICs; but these TICs recover rapidly following termination of macrophage inhibition. This is consistent with previous studies suggesting that melanoma cells can undergo phenotype switch3 and given the right environment, that all melanoma cells may have the potential to initiate tumors in vivo.4 The flexibility of melanoma TICs and their ability to exploit micro-environmental factors poses additional challenges for therapeutic development. Indeed, we observed that melanoma TICs in young mice do not respond to TAM stimulation; but acquire TAM-responsiveness following temozolamide treatment.2 Furthermore, chemotherapy recruits TAMs into the tumors and these TAMs protect melanoma TICs from the effects of chemotherapy. Thus chemotherapy promotes the development of new tumor clones that adapt to the changing environment and contribute to relapse.Surgical resection is the standard first line therapy of many types of cancer. Too frequently however, patients succumb to postoperative relapse and accelerated metastatic growth. While many strategies are currently developed to boost the immune system in order to assist tumor clearance after surgery, few have met clinical success; the potential tumor-promoting role of macrophages in postsurgical events may have been overlooked. Recently we demonstrated that macrophages play a central role in postoperative relapse at the primary site and in surgery-induced metastatic outgrowth.5 Similar to chemotherapy, surgery promotes the development of TAM-responsive TICs. Accordingly, macrophage depletion following surgery significantly reduced postsurgical recurrence and metastatic outgrowth compared to surgery alone. Recent studies in both mice6 and patients7 have demonstrated beneficial effects of macrophage depletion as a monotherapy. Together, our data highlights the feasibility of macrophage depletion as an adjuvant to surgical resection of solid tumors and to chemotherapy.However, before macrophage depletion can become feasible in the clinic, several issues will need to be resolved. These include effects on the physiological functions of macrophages such as wound healing. We did not observe any wound healing defects when we combined macrophage depletion with surgical resection, likely because our treatment did not deplete all macrophages but rather reduced the increase in macrophage infiltration brought on by surgery. Thus the level of macrophage depletion is an important factor to consider. Macrophage depletion may also affect other components of the immune system, such as neutrophils and T cells which will also alter the course of disease. Finally, one needs to optimize the method of depletion, using antibodies or small molecules, inhibiting macrophage function or their recruitment; the time window and length of therapy. Despite these challenges, we believe that targeting macrophages and the mechanisms of their interaction with different tumor cell populations will provide a new arsenal of treatments against cancer.