ABSTRACT: Metastasis to the bone represents a frequent complication of visceral cancers, most commonly in patients with advanced breast, prostate, and lung cancer. More than 50% of patients with advanced breast or prostate cancer have identifiable bone metastasis, and 30% to 40% of patients with non–small-cell lung cancer ultimately develop metastases to bone. Most tumors preferentially metastasize to the axial skeleton, targeting the vertebrae, pelvis, proximal ends of long bones, and skull. Skeletal complications as a result of these metastases are widely recognized to increase mortality and decrease quality of life—specifically the loss of mobility, independence, and social functioning of a patient. Advances in understanding the mechanisms of metastasis to bone, the resulting physiologic disturbances that take place, screening, diagnosis, and availability of better treatment options are advancing clinicians’ abilities to combat this devastating problem.
More than 50% of patients with advanced breast or prostate cancer have identifiable bone metastasis, and 30% to 40% of patients with non–small-cell lung cancer ultimately develop metastases to bone.[1,2] Most tumors preferentially metastasize to the axial skeleton, targeting the vertebrae, pelvis, proximal ends of long bones, and skull.[3] Skeletal complications as a result of these metastases are widely recognized to increase mortality and decrease quality of life—specifically the loss of mobility, independence, and social functioning of a patient.
Bone has several features that render it a target of circulating tumor cells. Stephen Paget first noted this in 1889 in his “seed and soil” hypothesis, whereby he postulated that the complimentary characteristics of a target organ and circulating tumor cells determines where tumors ultimately metastasize.[4,5] With bone, the high blood flow to the marrow serves to harbor and supply certain tumor cells with the necessary media in which to grow and multiply. Successful metastasis is largely dependent on immediate survival during migration, clonal expansion, and development of local blood supply. Additionally, tumor cells express adhesive molecules that allow them to bind to stromal and matrix elements of the bone.
Healthy bone undergoes constant remodeling, directed by the influence of endocrine hormones like parathyroid hormone (PTH) and 1,25-dihydroxyvitamin D (1,24-(OH)2D3), paracrine hormones, and cytokines. In the setting of metastasis, the structural integrity of bone is compromised due to disruptions in the normal balance of osteolytic and osteoblastic activity. The resulting lesions leave the bone weakened and prone to damage even under low-impact stress events. This commonly results in debilitating bone pain and can also lead to other skeletal-related events (SREs) such as pathologic fractures, bone deformation, leukoerythroblastic anemia, subsequent surgery or radiation, threatening hypercalcemia of malignancy, and spinal cord or nerve root compression.[6] Most primary research, however, reports on four of these SREs: radiation, surgery, fractures, and spinal cord compression.
Incidence of SREs and Subsequent SREs
A recent retrospective review found the incidence of bony metastasis to be 24% and 30%, respectively in American and Japanese patient populations with non–small-cell lung cancer (NSCLC). Of these 66% were noted to have skeletal lesions at the time of their initial staging on positron-emission tomography (PET)/computed tomography (CT) scan.[7]
![Figure 1: Distribution of skeletal-related events in 1-year retrospective study of a cohort of men with advanced prostate cancer metastasized to bone. Adapted from Lage et al [10].](http://www.cancernetwork.com/image/image_gallery?img_id=1927679&t=1313433385662 "Figure 1: Distribution of skeletal-related events in 1-year retrospective study of a cohort of men with advanced prostate cancer metastasized to bone. Adapted from Lage et al [10].")
Distribution of skeletal-related events in 1-year retrospective study of a cohort of men with advanced prostate cancer metastasized to bone. Adapted from Lage et al [10].
In the placebo arm of a randomized control trial, Rosen et al found that patients with lung carcinoma or other visceral tumors with metastasis to bone (excluding carcinomas of the breast and prostate) experience an average of 2.71 SREs/year.[8] The risk of additional SREs after the first also increases with each event. Other studies have noted that patients can experience up to four SREs per year, the cumulative incidence of which can have negative consequences on overall survival.[9] One analysis of insurance claims of patients with metastatic prostate cancer found that 22% of patients experienced more than one SRE with a mean time of 2.9 months between identification of the first type and the second type of SRE (if they were different, eg, surgery and pathologic fracture).[10] This same study also found that most patients experience radiation therapy, as shown in Figure 1, when they experience a SRE. Only 23.4% of SREs were found to be pathological fractures.
RANK Ligand Pathway
Factors Influencing Osteoclast and Osteoblast Activity
The increased risk of SREs in the setting of bony metastasis stems from a series of perturbations in the normal physiologic activity of bone remodeling. Normal bone is composed of two types: cortical bone, composed of hard, mineralized matrix; and trabecular or cancellous bone, which is less dense and more metabolically active. Cortical bone accounts for 85% of the total bone mass and can be found in the long bones of the appendicular skeleton. The other 15% makes up the cancellous bone that houses the red marrow, which contains stromal cells and hematopoietic stem cells that differentiate into osteoblasts and osteoclasts, respectively. Cancellous bone is found primarily in the vertebral bodies and pelvis. Osteoblasts are responsible for the deposition of new mineral matrix, creating new bone. Osteoclasts are responsible for the resorption of old bone in response to hormonal milieu. The mineralized matrix of bone contains immobilized cytokines and other growth factors that can directly influence neighboring cells—factors like tumor growth factor–beta (TGF-beta), insulin-like growth factor–2 (IGF-2), bone morphogenic proteins (BMPs), fibroblast growth factors (FGFs), and platelet-derived growth factor (PDGF).[11] Until bone is resorbed through osteoclastic activity, though, these cytokines remain fixed and nearby cells are not influenced by their hidden presence. A list of factors that regulate remodeling of bone is shown in Table 1.
The human body regulates calcium homeostasis and phosphate homeostasis in large part through parathyroid hormone (PTH), 1,25-dihydroxyvitamin D, and their action on bone, the GI tract, and the nephron. Calcitonin, a product of thyroid parafollicular C-cells, plays a small role in the activation of osteoblasts. In the bone, while osteoclasts oppose the actions of osteoblasts, the former cells, in fact, rely on the latter for cues in activity. Namely, osteoblast stimulation results in the release of receptor activator of RANK ligand. RANK ligand acts on its receptor, RANK, found on osteoclastic progenitor cells, to induce osteoclast formation and activity. At the same time, osteoblasts secrete osteoprotegerin (OPG), a decoy receptor of RANK ligand, effectively competing with RANK ligand to activate osteoclastic progenitor cells. The ratio of available RANK ligand and OPG, which is further influenced by local cytokines and growth factors, determines the net deposition or resorption of bone.
This depicts elements of normal bone physiology, namely the input of parathyroid hormone, 1,25-dihydroxyvitamin D, prostoglandins and cytokines in stimulation of osteoblasts to upregulate the expression of RANKL and release of osteoprotegerin (OPG), decoy receptors for RANKL that inhibit RANKL activity of corresponding receptors on osteoclastic progenitor cells. Osteoclasts will routinely resorb bone, releasing growth factors like IFG-II, TFGß, PDGF, BMPs that serve to restore homeostatic balance by stimulating bone deposition and remodeling. With a metastatic lesion, the cancer, begins a “vicious cycle” by secreting factors like PTHrP and other cytokines to stimulate an osteolytic lesion, effectively creating a positive feedback loop that leads to more destruction and weakening of bone. Calcium and other byproducts of the destruction of bone mineral matrix, like urinary N-telopeptide (uNTx) are also released in the process of osteoclastic activity.
A so-called “vicious cycle” is created when tumor cells synthesize and release osteolytic factors, including PTH-related protein (PTHrP), interleukin-6 (IL-6), and tumor necrosis factor–alpha (TNF-alpha), among many others. These stimulate osteoclasts, which in turn resorb bone and then release the fixed growth factors, which positively influence the growth of tumor cells. These growth factors close a positive feedback loop on the tumor cells by activating them, creating an endless loop of activity that drives an osteolytic lesion.[12,13] These pathways and their pathologic deviations are depicted in Figure 2 and discussed in greater detail in the paper by Murthy et al earlier in this supplement.
Types of Bone Metastasis
Metastatic lesions of the bone can be classified in one of three ways: osteolytic, osteoblastic, or mixed.[14] Most cancers result in osteolytic lesions (carcinoma of breast and lung). In vivo studies have shown that osteolysis is due to osteoclastic activity, and is not a direct effect of the cancer itself. Overall, osteolytic lesions result from greater osteoclast than osteoblast cellular activity, resulting in uncoupled bone resorption. That is, there is no negative feedback mechanism to halt the process.[15] Targeting the various intersections within the positive feedback loop with appropriate therapies can clinically prevent further degradation of the bone, preserving function and preventing or delaying SREs.[16]
Prostate cancer (and 15% to 20% of breast cancers) results in osteoblastic lesions of the bone. In these lesions, the balance has been thought to shift to more bone deposition than destruction, as shown with correlation of levels of alkaline phosphatase and osteocalcin, both markers of osteoblastic activity. However, the bone that is deposited is of poor quality and often leads to severe bone pain and still puts the patient at risk for pathologic fractures and other subsequent SREs.
Research implicates the activity of Endothelin-1 (ET-1) as a mediator of this activity through both paracrine and autocrine activity on osteoblast cells. Prostate cancer and some breast cancer cells secrete high levels of ET-1, which in turn stimulate osteoblast cells.[17] An in vitro analysis found that an ET-1 receptor antagonist reduced osteoblastic activity.[18] The microenvironment of bone may further enhance ET-1 activity through the stimulation of prostate cancer cell proliferation and potentiating the effect of other local growth factors.[19]
That there exist two types of bony metastatic lesions is likely an oversimplification, as normal physiology dictates a carefully choreographed balance of activity from osteoclasts and osteoblasts. Perturbation of this healthy model does not likely inactivate one aspect of normal bone homeostasis, but rather results in an over-preponderance of bone deposition or destruction. This view was validated by a mouse model examining the activity of osteoblastic xenograft breast cancers.[20]
Researchers initially saw an increase in osteolytic activity followed by a resurgence in osteoblastic deposition of bone. Similarly, osteolytic lesions often result in new (albeit ineffective) bone deposition in response to the loss of healthy bone. As such, even in osteoblastic lesions specific to prostate cancer, osteolytic activity is comparable to that observed in breast cancer and multiple myeloma. Because osteolysis plays an important role in both lesions, it is believed that pharmacologic targeting—especially by bisphosphonates, which are potent inhibitors of osteoclasts—can serve to mitigate the horrendous clinical effects of such metastatic lesions.
