B cells and macrophages in cancer: yin and yang
Nature Medicine 17, 285–286 (2011)
Inflammation is an important component of the tumor microenvironment; however, the mechanisms through which immune cells might promote tumorigenesis are unclear. A recent study indicates that B cells and antibodies have a key role in orchestrating macrophage-driven, tumor-promoting inflammation, suggesting that modulating the pathways involved might be of therapeutic benefit in cancers driven by chronic inflammation.
Cancer-related inflammation is characterized by the recruitment of cells of the monocyte-macrophage lineage to tumor tissues1, 2, 3. Cancer-causing genetic events, such as oncogene activation and inactivation of tumor suppressor genes, elicit leukocyte recruitment and the formation of an inflammatory microenvironment. In turn, myelomonocytic cells can affect virtually all steps of carcinogenesis, including genetic instability and metastasis.
In progressing tumors, tumor-associated macrophages (TAMs) generally express an M2-like phenotype3, which is characterized by low interleukin-2 (IL-2) expression, high IL-10 expression and low tumoricidal activity, and promotes tissue remodeling and angiogenesis (Fig. 1). The functional orientation of TAMs is orchestrated by tumor-derived and host-derived signals3. In most human tumors, TAM infiltration is associated with poor prognosis, as seen in Hodgkin's disease4.
Figure 1: B cell–mediated orchestration of tumor-associated macrophages.
Various pathways orchestrate the protumor function of M2-like myelomonocytic cells in tumors. Andreu et al.5 showed that B cells produce antibodies that can form immunocomplexes (ICs) that orient TAMs to promote cancer via Fcγ receptors on the surface of TAMs. T helper type 2 (TH2)-derived IL-4, tumor-associated fibroblasts (TAFs) and tumor cells themselves provide alternative or complementary pathways that can skew macrophages toward an M2-like protumor
A recent study by Andreu et al.5 using a mouse model of human papilloma virus-driven multistage squamous epithelium carcinogenesis has provided insights into the complexity of the pathways involved in skewing mononuclear phagocyte function in this type of cancer. B lymphocytes were found to orchestrate TAM functions by 'remote control' by producing antibodies that interact with and activate Fcγ receptors on both tumor-resident and recruited myeloid cells (Fig. 1). These results suggest that B cell– and Fcγ receptor–mediated pathways might be therapeutically targeted in individuals with chronic inflammatory disease who are at risk of developing cancer or even in established tumors.
Andreu et al.5 showed that during carcinogenesis, with the help of CD4+ T cells, B cells produced antibodies directed against extracellular matrix components at the tumor site. Fcγ receptor–mediated recognition of immune complexes containing these antibodies, most likely in concert with myeloid differentiation factor-88–dependent signaling, led to mast cell–dependent angiogenesis and recruitment of mononuclear phagocytes. In this tumor model, TAM-mediated enhancement of carcinogenesis is complement independent, but complement components can drive myelomonocytic cell recruitment in other tumors3, 6. In the tumors studied by Andreu et al.5, mononuclear phagocyte-dependent tumor promotion is mediated by a mature, M2-like macrophage population rather than by immature cells (the myeloid-derived suppressor cells) in the myelomonocytic differentiation pathway. The immune complex-conditioned TAMs can in turn participate in yet another pathway of tumor promotion mediated by fibroblasts7 (Fig. 1). In this pathway, via macrophage-derived IL-1, carcinoma cells direct fibroblasts to recruit more macrophages and promote angiogenesis.
Functionally skewed cells of the myelomonocytic differentiation pathway are a common denominator of inflammation-promoted carcinogenesis, although the cell types and mediators involved can differ considerably. For instance, in a mouse model of breast carcinogenesis and metastasis, the culprit of M2 polarization of TAMs and tumor promotion was T helper type 2 cell–derived IL-4 (ref. 8) (Fig. 1). Similarly, the mechanisms of B cell–mediated tumor promotion need not be restricted to antibody production by these cells3, 9 (T. Schioppa and F. Balkwill, Queen Mary University of London, personal communication). B cells have also been shown to drive M2-like polarization of macrophages and to promote the growth of transplanted B16 melanomas9. In this case, the promoter of M2 macrophage polarization was a B cell subset (B-1 cells) characterized by constitutive IL-10 production. In prostate cancer, androgen ablation results in tissue damage and leukocyte recruitment, including B cells. In this example, B cells produced lymphotoxin, which promoted the hormone-independent growth of prostate cancer10.
Antibodies are also part of the anticancer therapeutic armamentarium, and there is strong evidence that macrophages and Fcγ receptors are key to their antitumor activity, such as for the effects of the CD20-specific monoclonal antibody rituximab. Treatment of a xenograft lymphoma model with rituximab resulted in leukocyte recruitment mediated by the chemokines CCL3 and CCL4 and tumor eradication11. However, combined treatment with a CCL3 antagonist and rituximab compromised the antitumor activity of rituximab, and an absence of CCL3 signaling led to macrophage depletion, suggesting that macrophages mediate the antitumor effects of this antibody, as also indicated by clodronate-mediated macrophage depletion11. Interestingly, rituximab elicits an M2-like phenotype in macrophages, leading to the increased phagocytosis of mouse lymphoma cells12. However, in people with lymphoma, the combination of rituximab with high-dose chemotherapy and autograft with peripheral blood progenitor cells was associated with an increased frequency of solid tumors in a large 20-year retrospective follow-up study13. Thus, the interplay between macrophages and B cells is complex, and, depending on the context, it can result in promotion of carcinogenesis5 or antitumor activity11.
The results reported in these studies raise issues related to the diversity of the cancer-related inflammatory response and the design of therapeutic strategies to target this response. In different tissues and tumors, the pathways that orchestrate cancer-related inflammation can differ considerably, with a pivotal role of B cells and antibodies in squamous carcinogenesis5, 8. Strategies targeting B cells and IL-1 are currently in clinical use and may provide information relevant for defining the diversity of cancer-related inflammatory responses in humans, as well as innovative therapeutic strategies14. Myelomonocytic cells come in various types in different tumor contexts2, 3, but mononuclear phagocytes emerge as an essential common constituent of cancer-related inflammation. The identification of the various cellular and molecular pathways that participate in inflammation in different human cancers will be required to translate a better understanding of cancer-related inflammation to the bedside.