Skeletal metastases are a common event in breast cancer, responsible for profound patient morbidity and representing a significant public health burden. While the precise epidemiology of bone metastases in breast cancer is elusive, they affect over 7% of all women diagnosed with breast cancer in the United States (1). Bone is the initial site of metastasis for 47% of women with relapsed breast cancer (2) and the only site for 20% of women with metastases (3). The incidence may vary by subtype: among patients whose initial metastasis is to the bone, 82% will express either estrogen receptor (ER) or progesterone receptor (PR). Nonetheless, these events are common in all subtypes, noted in 69% of women who die from breast cancer (2). Importantly, the development of bone metastases is not a fatal event. Patients with skeletal metastases can have a prolonged survival, particularly when compared to patients with visceral involvement. One study reported a median survival of 24 months for patients with only skeletal metastases compared to 3 months for patients with liver metastases. As our treatment for advanced breast cancer evolves, we are likely to see further improvements in survival, increasing the chances for patients to develop bone metastases during their lifetimes. Because of the high incidence of bone metastases and the potential for extended survival, the cumulative morbidity of skeletal involvement in breast cancer is high and the potential for impaired mobility and loss of independence highlights their clinical relevance.

The morbidity from bone metastases has been difficult to objectively quantify. While helpful, the use of pain scores and analgesic logs is subject to various potential biases. As a result, the impact of bone metastases is often quantified by the incidence of skeletal-related events (SRE). SRE definitions vary by study but operationally represent a composite of pathologic fractures, spinal cord compression with vertebral compression fracture, hypercalcemia of malignancy, requirement for radiation to bone, and necessity for surgical intervention for pathologic fractures or cord compression (4). Unfortunately, SRE are common in breast cancer. In the absence of bone-targeted therapy, the rate of pathologic fracture for patients with bone metastases from breast cancer is 35% (5) and up to 70% of patients with bone metastases will have at least one SRE by 2 years (6). SRE are associated with decreased survival in breast cancer. Patients with a pathologic fracture have a 32% increased risk of death compared to those without a fracture (7). The clinical significance of SRE, however, extends far beyond survival. SRE have been associated with significant declines in physical and functional well-being as well as overall quality of life (8). The resulting economic impact of bone metastases in breast cancer is staggering, with an estimated national cost burden of nearly $4.5 billion annually (1).

Insight into the pathophysiology of bone metastasis has facilitated recent therapeutic advances. The tropism of breast cancer cells to bone has yet to be fully explained. Transcriptome analyses of breast cancer cells predisposed to metastasize to the bone have revealed a gene expression profile that mimics osteoblasts in a process called osteomimetism. These changes may facilitate survival of these cells in the bone microenvironment (9). Under normal physiologic conditions, osteoclast differentiation is directed by osteoblasts via signaling in the receptor activator of NF-κB (RANK) ligand pathway (10). The interactions between these molecules results in normal bone remodeling: a balance between bone formation and resorption. In the setting of bone metastases, normal homeostasis is disrupted. Osteoclast activation is felt to be an early event in the development of bone metastases in breast cancer, often associated with lytic bone lesions. Dysfunction of osteoblasts has been implicated in blastic bone metastases and is relevant in breast cancer, where mixed lesions are frequently encountered.

This dysfunction may be mediated by the RANK ligand pathway, an emerging target in the treatment of bone metastases. An increase in bone resorption can compromise the integrity of the affected bone, leaving it vulnerable to the development of SRE. This process can be mediated by several factors. One contributing factor is an increase in osteoclast precursors in the presence of tumor cells, a shift potentially mediated by tumor necrosis factor α(11). Another mechanism is activation of osteoclasts by breast cancer cells, which can upregulate RANK ligand and inhibit osteoprotegerin, which normally inhibits excessive osteoclast activation. Once activated, bone resorption can elicit a chemotactic response from tumor cells, cyclically propagating the process (12). Agents targeting this pathway, however, have provided inroads in stopping this cycle and preventing some of the morbidity associated with skeletal metastases.

Bisphosphonates are a class of medications commonly used in patients with bone metastases from breast cancer. Early-generation bisphosphonates such as clodronate do not contain nitrogen. These agents are metabolized to adenosine triphosphate (ATP) analogues that cannot be hydrolyzed, interrupting multiple cellular processes including bone remodeling (13). Nitrogen-containing bisphosphonates, including pamidronate, ibandronate, and zoledronic acid, accumulate in the bone microenvironment and are released during bone resorption. They are then internalized by osteoclasts where they affect their activity and survival. One target for nitrogen-containing bisphosphonates is farnesylpyrophosphate synthase (FPPS), an enzyme in the mevalonate pathway that is responsible for the prenylation of small guanosine triphosphatases (GTPases), which are critical for osteoclast survival (14). This indirect effect of nitrogen-containing bisphosphonates is augmented by a direct effect related to production of triphosphoricacid 1-adenosin-50-yl ester 3-(3-methylbut-3-enyl) ester (ApppI), an ATP analogue that induces osteoclast apoptosis (15). The effect of bisphosphonates may extend beyond the osteoclast. Preclinical studies have shown that zoledronic acid inhibits angiogenesis in several in vitro and in vivo models (16). In patients receiving zoledronic acid, there is an early and sustained reduction in circulating levels of vascular endothelial growth factor (VEGF), a cytokine involved in angiogenesis (17). Zoledronic acid also modulates migration and adhesion of endothelial cells by targeting integrins and sensitizing cells to tumor necrosis factor (TNF)-induced programmed cell death (18). In addition, nitrogen-containing bisphosphonates prevent adhesion of cancer cells to bone, which may interfere with tumor seeding and invasion (13). Nitrogen-containing bisphosphonates may also induce cell death by activating cytotoxic T-cells in the host, though the clinical role of this effect awaits further clarification (19).

Agents targeting the RANK ligand pathway have also demonstrated activity, primarily through inhibition of osteoclast function to suppress bone resorption (20). Denosumab is a humanized monoclonal antibody targeting RANK ligand that has recently garnered FDA approval in this setting (21). Similar to zoledronic acid, the activity of denosumab is mediated by several mechanisms, including a potentially similar antiangiogenic effect (22). Multiple randomized studies have shown a decrease in SRE with the use of bisphosphonates and denosumab, which are outlined next.