Abstract #DDT02-2: MCS110: a monoclonal antibody with potent neutralizing activity against macrophage colony-stimulating factor for the treatment of tumor-induced osteolysis

Macrophage colony-stimulating factor (M-CSF), also known as colony-stimulating factor-1 (CSF-1), is a key driver of the differentiation of osteoclasts from precursor cells of the monocytic lineage. In cancer patients with bone metastases, it has been hypothesized that secretion of M-CSF by invasive tumor cells or by tumor stroma leads to aberrant osteoclastogenesis, resulting in an increase in osteoclast cell number, excessive bone degradation, and skeletal-related events, including bone pain and fractures. MCS110 is a high-affinity Human Engineered\#8482; anti-M-CSF monoclonal antibody (mAb) that blocks the ability of M-CSF to drive proliferation in responsive cells. MCS110 also inhibits M-CSF-stimulated phosphorylation of the M-CSF receptor, CSF-1R. In addition, MCS110 completely blocked the osteoclast differentiation process in an in vitro osteoclastogenesis assay using CD34 + human bone marrow cells stimulated with M-CSF and the receptor activator of NF-kappa-B ligand (RANKL). The mechanism of action of MCS110 was evaluated in vivo by treating cynomolgus monkeys with the antibody, which resulted in sustained loss of serum M-CSF bioactivity. The animals also exhibited decreased bone resorption, as reflected by 40-70% decreases in serum and urine markers of bone degradation, as well as reduced osteoclast numbers detected by histology. To investigate the role of M-CSF in bone destruction associated with metastatic cancer, the effects of M-CSF neutralization were examined in three murine xenograft tumor-induced osteolysis (TIO) models. These models involved intratibial injection of human tumor cell lines in immune-compromised mice, followed by treatment of the mice with anti-M-CSF antibodies and/or other agents. The incidence and severity of osteolytic lesions formed in the tibia were scored in a blinded fashion from x-ray images taken at the end of each study. Because MCS110 is primate-specific, a combination of surrogate anti-mouse and anti-human antibodies (to block host-derived and xenograft-derived M-CSF, respectively) was used in these studies. In TIO models of human breast cancer (MDA-MB-231Luc) and prostate cancer (PC-3MLuc), the combined anti-M-CSF antibodies were highly effective in reducing both the incidence and severity of the bone lesions by as much as 70% compared to the IgG control antibodies. In addition, concordant reduction of tartrate-resistant acid phosphatase 5B (TRAP5B), a marker of osteoclast abundance, was observed in both models. Assessment of the level of luciferase-expressing cells in the tibias using bioluminescent imaging indicated that treatment of the mice with the anti-M-CSF antibodies did not have a detectable effect on tumor burden in these models. In a mixed osteolytic/osteoblastic xenograft model of metastatic bladder cancer (TSU-Pr1), the anti-M-CSF antibodies significantly reduced the severity and incidence of both the osteolytic and osteoblastic lesions. Anti-M-CSF antibodies also showed enhanced activity in these models when combined with the bisphosphonate zoledronic acid (Zometa®), compared to either single-agent treatment. Taken together, these results have provided the basis for the initiation of clinical trials of MCS110 in patients with bone metastases, to explore whether MCS110 will inhibit aberrant osteoclastogenesis and skeletal-related events in these patients. Soluble forms of receptor tyrosine kinases have been identified in cancer patient sera and have been proposed as biomarkers for targeted drug therapies. We therefore tested for the presence and regulation of soluble forms of CSF-1R. Analysis of conditioned medium samples taken from the in vitro osteoclastogenesis assay revealed that a soluble, presumably shed version of CSF-1R is released from the cells during this process. A circulating form of CSF-1R was also detected in human serum and plasma samples. If the circulating form of CSF-1R in vivo is generated in large part during the differentiation of osteoclasts and other cells of the monocytic lineage, we hypothesized that treatment with neutralizing antibodies against M-CSF should result in a decrease in the circulating soluble CSF-1R concentration. In contrast, treatment with bisphosphonates should have no effect on the soluble CSF-1R concentration, since these agents primarily affect the osteoclasts themselves, rather than their differentiation. To test the hypothesis, a mouse CSF-1R ELISA assay was developed and used to measure the level of soluble CSF-1R in serum samples taken from animals in the TIO models described above. Significant decreases in the circulating CSF-1R levels were seen in the mice treated with the anti-M-CSF mAbs ± zoledronic acid, but not in those animals treated with zoledronic acid alone. The circulating CSF-1R level was also reduced in cynomolgus monkeys treated with MCS110, and this decrease correlated with the observed reduction in the serum concentration of the bone resorption marker NTx (cross-linked N-telopeptide of type I collagen). Thus, the soluble form of CSF-1R may prove useful in the clinic as a biomarker of MCS110 action, particularly in patients whose markers of bone turnover are already depressed due to concomitant or prior treatment with bisphosphonates. Citation Information: In: Proc Am Assoc Cancer Res; 2009 Apr 18-22; Denver, CO. Philadelphia (PA): AACR; 2009. Abstract nr DDT02-2.