Macrophages in the liver prevent metastasis by efficiently eliminating circulating tumor cells after monoclonal antibody immunotherapy.

Monoclonal antibodies (mAbs) are increasingly being used to treat cancer. In response to mAb therapy, we have identified macrophages in the liver as major effector cells removing circulating tumor cells via antibody-dependent phagocytosis, an immune cell-mediated process that prevented liver metastasis. This discovery extends our understanding of the mechanisms of mAb therapy, and may help to optimize mAb-based anticancer therapeutics.

Introduction

The use of antitumor monoclonal antibodies (mAbs) has increased dramatically in the last decade and is now a mainstream strategy to treat cancer patients. Nonetheless, their mode of action has yet to be completely resolved. mAbs exert direct effects on tumor cells such as growth inhibition or apoptosis induction. Indirect anticancer properties include mAb-mediated activation of the complement pathway leading to complement-dependent cytotoxicity as well as mAb-solicited recruitment of immune effector cells that express high affinity immunoglobulin-Fc receptors. Treatment with antitumor mAbs has been shown to be ineffective in mice lacking 1 or more of the activating Fcγ receptors, supporting the importance of Fc receptor-mediated mechanism(s) of action for in vivo therapeutic efficacy of antibody based immunotherapy.

With live cell imaging and intravital microscopy we recently demonstrated that mAb therapy potently induces phagocytosis of tumor cells by macrophages manifesting in the elimination of circulating tumor cells by Kupffer cells (liver macrophages) and preventing liver metastases. Whereas in the absence of mAbs, Kupffer cells interacted with and sampled portions of tumor cells, antibody-dependent cellular phagocytosis (ADCP) was required for complete tumor cell eradication (Fig. 1). ADCP was found to be dependent upon FcγRI and FcγRIV, consistent with previous studies in which we showed that either FcγRI or FcγRIV was required to prevent outgrowth of liver metastases after mAb therapy.

Figure 1. Kuppfer cells in the liver eliminate circulating tumor cells by antibody-dependent cellular phagocytosis after treatment with antitumor monoclonal antibodies. Left: In the absence of antitumor mAbs, Kupffer cells (blue) are able to interact with tumor cells (red), and sample portions. However, this sampling is insufficient to prevent outgrowth of liver metastases. Right: After treatment with antitumor monoclonal antibodies (mAbs), Kupffer cells rapidly and efficiently phagocytose tumor cells leading to intracellular cancer cell degradation in acidified lysosomes and preventing the development of liver metastases.

ADCP was associated with the generation of phagolysosomes within macrophages that were rapidly acidified. However, intracellular degradation of tumor cells was found to be a slower process both in vitro and in vivo. Production of reactive oxygen species (ROS) and nitrogen species were proposed as major cytotoxic mechanisms implemented by macrophages. However, even though ADCP stimulated the generation of ROS, neither ADCP, nor acidification of phagolysosomes and resultant breakdown of tumor cells was found to be dependent on ROS. Thus, intracellular digestion in lysosomes is the most likely mechanism by which macrophages kill tumor cells in the process of mAb-mediated phagocytosis.

Macrophages may play a crucial role in the therapeutic success of anti-CD20 mAb therapy in patients with B cell malignancies. Supporting this premise, macrophage depletion abrogated the ability of anti-CD20 mAbs to eliminate lymphoma cells in an experimental model. Another recent study by Montalvao et al.showed that Kupffer cells trapped circulating normal and malignant B cells in the liver after anti-CD20 mAb therapy, and eliminated them through ADCP, independently confirming our findings. This is most likely due to the convenient localization of Kupffer cells in the vasculature, enabling easy access to both mAbs and circulating tumor cells. Interestingly, clinical responses after treatment with the anti-CD20 mAb rituximab were correlated with polymorphisms in human FcγRIIa and FcγRIIIa (FcγRIIa-131H/R and FcγRIIIa-158V/F) that affect affinity for IgG. Whereas both natural killer cells and macrophages express FcγRIIIa, only macrophages express FcγRIIa, strongly supporting a role for macrophages as effector cells in the depletion of B lymphoma cells after anti-CD20 mAb treatment of cancer patients.

It is currently unclear whether macrophages contribute to tumor cell killing after mAb therapy for the treatment of solid malignancies. Homozygosity for FcγRIIa-131H has been associated with stronger antitumor responses and progression-free survival when patients afflicted with metastasized breast cancer were treated with anti-HER2 mAbs (trastuzumab), findings supporting an anticancer role for macrophages. Additionally, macrophages isolated from breast carcinomas in mice have been found to be capable of ADCP. Furthermore, antitumor mAb therapy was reported to be less successful in preventing breast carcinoma outgrowth and metastasis after depletion of macrophages, suggesting that macrophages may be involved as effector cells following mAb therapy of breast cancer.

Polymorphisms in FcγRIIa-131H/R and FcγRIIIa-158V/F have been further correlated with clinical responses of patients with colorectal cancer after treatment with the anti-epidermal growth factor receptor (EGFR) mAb cetuximab. However, we found that mAb therapy was ineffective in treating existing liver micro-metastases, as Kupffer cells proved stationary and were not recruited into micro-metastases. Thus, these results argue against an important role for Kupffer cells in mAb therapy once liver metastases have been established, a premise supported by current standard clinical practice. Anti-EGFR mAbs are only indicated for treatment of metastatic colorectal cancers with wild type RAS, as this therapy is ineffective when tumors harbor a RAS mutation. This further supports the notion that the direct effects of anti-EGFR mAbs (such as growth inhibition) are more important than Fcγ receptor-mediated effector mechanisms in the treatment of established and already metastatic colorectal tumors.

Nevertheless, our results may hold great promise for the treatment of colorectal cancer patients. Approximately 1.2 million patients worldwide are diagnosed with this devastating disease each year, with an estimated annual death rate of 600 000 patients. Surgical removal of the primary tumor is currently the only therapy that can provide long-term disease-free survival, such that surgical resection is an absolute necessity. Unfortunately, many patients still develop metastases, despite successful resection of the primary carcinoma at a time when metastatic disease had not yet manifested itself. However, disseminated circulating tumor cells can be detected at the time of surgery in the majority of patients. Moreover, the presence of circulating tumor cells is associated with disease progression and poor survival, regardless of whether following surgical removal of the primary tumor or, whenever possible, liver metastases.

Paradoxically, even though surgical removal of primary colorectal cancer is a prerequisite, resection of the tumor contributes to the risk of developing liver metastases. Evidence for this unfortunate occurrence can be found in our own work showing that surgery promotes adherence of circulating tumor cells to the liver. We now show that treatment with antitumor mAbs potently induces ADCP of tumor cells by Kupffer cells, thereby preventing the development of liver metastases. As such, we propose that patients at risk of developing (recurrent) liver metastases—including patients undergoing resection of primary colorectal cancer or resectable liver metastases—will profoundly benefit from pre-operative mAb adjuvant therapy.

In conclusion, we have shown that Kupffer cells are critical for the elimination of circulating tumor cells via ADCP following antitumor mAb-mediated immunotherapy. As such, promoting recruitment of macrophages as effector cells in mAb-based strategies may enhance therapeutic efficacy. This may be particularly important for patients with hematological malignancies, as well as those with circulating tumor cells at the time of elective cancer surgery, in order to prevent the development of post-surgical liver metastases.