Prostate Cancer (PC) represents a major health issue in men with an estimated 240,000 new cases and almost 30,000 deaths annually in the United States [
28]. Although surgery and radiation therapy has cured many patients with PC, more than one third of the patients will eventually progress and develop advanced disease, for which ADT is the standard of care. Unfortunately, most patients who undergo chemical or surgical castration will eventually progress and develop CRPC [
29,
30]. There have been numerous studies attempting to understand the mechanisms and processes for the development of CRPC, some of which include ligand-independent AR activation, AR mutations, amplification and selection of pre-existing androgen independent cells that are resistant to apoptosis [
31,
32].
Bone is the most common site of metastases in men with advanced PC. Historical data shows that the overwhelming majority (more than 80%) of PC patients eventually develop bone disease [
6,
7,
33]. Cancer metastases in the bone usually lead to disturbance of bone formation and resorption, resulting in osteolytic and/or osteoblastic lesions [
34‐
36]. It remains largely unclear what factors determine whether a particular bone metastasis is osteolytic, osteoblastic or of mixed phenotype, although some studies suggested that different types of cancer have the propensity to secrete more osteoblastic or osteolytic factors [
37]. In this study, we observed different bone phenotypes between castrated and intact host mice intra-tibially implanted with VCaP-Luc cell line. Although this cell line is classified as a castration-resistant prostate cancer line, its growth rate was reduced in castrated mice compared with intact mice, in both subcutaneous and intra-tibial growth conditions. This difference may lead to difference in growth characteristics of the cells and in the type of paracrine factors that affect the activation of osteoblasts and osteoclasts. Regardless of the phenotype, bone metastases usually lead to compromised bone integrity and strength, and as a result, increased incidence of fractures, spinal cord compression and severe pain [
38,
39]. These are not only quality of life issues, but also directly worsen the survival span of the patients. Thus, effective prevention and management of bone metastases is critical for reducing PC related morbidity and mortality. Until recently, there have been limited choices for therapeutic intervention of metastasis-induced bone loss. Although zoledronic acid has been used to prevent bone loss, its long term benefit is not well defined [
40]. Newer agents such as Denosumab [
41] and cabozantinib [
42] have shown clinical benefit in delaying the occurrence of skeletal-related events (SRE) in PC patients.
Crizotinib is a potent and selective inhibitor of c-MET and several related tyrosine kinases. C-MET and its ligand, hepatocyte growth factor (HGF), are involved in cell proliferation, differentiation, motility and survival [
43]. Crizotinib exhibited a dose-dependent anti-tumor effects by inhibition of tumor cell proliferation, micro vessel density and angiogenesis [
44]. Tumor effects due to in prostate cancer, c-MET expression is elevated in bone metastases compared with lymph node metastases or primary tumors [
19], and in tumors in castrated patients compared with non-castrated patients [
45]. In addition, c-MET pathway is believed to play a role in proliferation, differentiation and migration of osteoblasts and osteoclasts [
46,
47]. The c-MET ligand, HGF can be secreted by multiple cell types including osteoclasts and mesenchymal cells, potentially forming autocrine and paracrine regulation of bone remodeling [
46]. Axitinib the other kinase inhibitor used in this study is a potent and selective inhibitor of angiogenesis targeting VEGFR2. Elevated levels of circulating VEGF have been correlated with poor prognosis in PC patients [
48]. VEGF and its receptors are shown to be expressed in osteoblasts and osteoclasts [
49], exerting multiple functions through autocrine and paracrine activation of the VEGFR pathway. Taking into account the function of c-MET and VEGFR in PC metastasis, we hypothesized that blocking both pathways simultaneously by combining two targeted agents may offer substantial benefit to PC patients. Our data showed that axitinib alone or axitinib and crizotinib in combination reduced the tumor burden in both intact (androgen positive) and castrated (androgen negative) mice models (Figure
3A and
3C). Interestingly only the axitinib and crizotinib combination showed improvements of bone volume (BV/TV ratio) in both models (Figure
4B and
5B). This suggested that combination therapy had clear benefit on tumor burden and bone volume in both intact (androgen positive) and castrated (androgen negative) mice models of PC. The combination can also be achieved by a single agent with dual or multiple specificities as indicated by cabozantinib, a potent inhibitor of receptor tyrosine kinases including MET, VEGFR2, AXL, FLT-3, KIT, and RET [
9]. In fact, a phase II adaptive, randomized discontinuation trial of cabozantinib in patients with metastatic CRPC [
42] has generated encouraging results showing clear improvements in bone scans in 68% of evaluable patients. On the other hand, co-administration of two separate compounds (such as crizotinib and axitinib) would allow flexibility when the dosage of each compound or pathway needs modification.