Emerging preclinical data in mouse models of breast cancer (BC) have implicated granulocyte colony-stimulating factor (G-CSF) in driving the development of metastatic disease. G-CSF expression has been demonstrated in a number of human cancers and is associated with poor prognosis. Neutrophils are the major G-CSF receptor (G-CSFR) expressing immune population. In patients, a high neutrophil to lymphocyte (NLR) ratio is associated with poor prognosis. Currently there are no neutrophil-targeted therapies used clinically for BC, despite reports that the immunosuppressive functions of these cells not only drive metastasis in mice, but could also be partially responsible for patients failing to respond to checkpoint inhibitors.
Using the 4T1 mammary tumour model, we have demonstrated that inhibition of G-CSFR can prevent metastasis without causing neutropenia in mice. By varying the timing of anti-G-CSFR therapy in a model encompassing primary tumour resection, we have demonstrated that early neoadjuvant treatment is required to inhibit metastasis. Reduced metastasis with G-CSFR blockade was associated with altered natural killer (NK) cell maturation in lung, while shRNA-mediated knockdown of G-CSF in 4T1 tumour cells resulted in reduced metastasis and decreased expression of the immunosuppressive gene Nos2 in neutrophils. Thus, we propose that G-CSF confers immunosuppressive activity to neutrophils against anti-tumour immune cells, thereby enabling metastasis. High PD-L1 expression was maintained on myeloid and tumour cells during G-CSFR blockade, suggesting combination with anti-PD-L1 may further increase the effectiveness of this therapy.
Our results reveal a contribution of G-CSF instructed neutrophils to metastasis. Early targeting of neutrophils with anti-G-CSFR therapy may aid in the prevention of metastatic disease or enhance immunotherapy efficacy in BC patients. As G-CSF is administered as a supportive therapy to prevent neutropenia in patients undergoing myelosuppressive chemotherapy, we recommend adoption of strict clinical guidelines, to avoid the potential for G-CSF-associated metastasis as observed in our preclinical studies.