Advances in
Bone Cancer Treatment:

Preventing Metastasis and Bone Loss

Anticancer Effects of Bone-Targeted Agents

Antitumor Actions of Bisphosphonates

Bisphosphonates are extensively used to treat cancer-induced bone disease in a range of solid tumors and multiple myeloma, where they reduce the incidence of SREs and improve patients' quality of life.^97-99 Ongoing clinical trials of breast cancer patients without confirmed bone involvement (AZURE, SWOG S0307, and NSABP B34) are expected to demonstrate whether adjuvant bisphosphonate therapy reduces the incidence of future bone metastases.^71,92 A key question is whether bisphosphonates contribute to reducing tumor growth, in addition to improving skeletal health. In support of this hypothesis, the antitumor effects of a range of bisphosphonates have been reported in different cancer cell types in vitro.^100,101 In vivo studies have shown a reduction in tumor burden in bone following bisphosphonate treatment in several model systems, including breast and prostate cancer, osteosarcoma, and multiple myeloma.^102,103 However, the majority of in vivo studies have been designed to investigate the effects of bisphosphonates on cancer-induced bone disease, not tumor burden, and have therefore utilized models of end-stage disease. It is possible that at this stage the tumor has reached a size where it is capable of autocrine growth, no longer depending on the surrounding microenvironment for expansion. In addition, in these studies very high doses and frequent administration of bisphosphonates were used; hence, the clinical relevance of these data has been the subject of considerable debate.¹⁰⁴

Bisphosphonate Mechanisms of Action

Bisphosphonates have a high affinity to bone, and they are rapidly cleared from the circulation. Their main targets are osteoclasts, which are responsible for bone resorption. Bisphosphonates can be divided into two categories: nitrogen-containing bisphosphonates like zoledronic acid, and non-nitrogen-containing (simple) bisphosphonates such as clodronate that have distinct mechanisms of action. Whereas nitrogen-containing bisphosphonates induce osteoclast apoptosis through inhibition of enzymes responsible for correct intracellular localization of key proteins, the simple bisphosphonates are metabolized into cytotoxic ATP analogues. Both classes of bisphosphonates ultimately cause osteoclast death and thereby reduce cancer-induced bone lesions.

Although we have a clear understanding of how they inhibit bone resorption, the exact cellular and molecular mechanisms responsible for the proposed antitumor effects of bisphosphonates remain to be elucidated. One major difficulty in pinpointing these effects is distinguishing between direct effects on cancer cells and those mediated through inhibition of bone resorption and the resulting reduction in the levels of bone-derived tumor growth factors. Bone matrix contains a range of different growth factors and other molecules that are released during bone remodeling, which help make bone a fertile soil for metastatic tumor growth.¹⁰⁵ An additional potential mechanism whereby bisphosphonates may affect tumor growth is through interfering with trafficking of bone marrow precursor cells to peripheral tumors. Most likely a combination of both direct and indirect effects contribute to reduced tumor growth caused by bisphosphonates, depending on the concentrations of drugs in different tissues. A key question is whether a low dose of bisphosphonates present in the circulation for a limited period of time following clinical administration is sufficient to elicit the antitumor effects demonstrated in model systems. There is currently no widely available method for measuring bisphosphonate concentrations in different tissues in vivo, and available surrogate markers are not sufficiently sensitive. As a result, we have little direct evidence that the reported antitumor effects are caused by bisphosphonates directly binding to molecular targets in peripheral (extraskeletal) tumors.

Effects on Tumors in Bone

Bisphosphonates are potent inhibitors of osteoclast activity, and numerous in vivo studies have reported that administration of these drugs causes a significant reduction in the number and extent of cancer-induced bone lesions. Positive effects have been reported in models of breast, prostate, myeloma, osteosarcoma, and other cancers.^102-104 Despite causing a very significant reduction in bone resorption, bisphosphonate treatment results in only a limited reduction in tumor burden in bone. In several studies bisphosphonate treatment has caused tumors to expand to bone-associated soft tissue. This may be a result of the high preservation of trabecular bone, limiting the available space for expansion of tumors inside the bone marrow cavity. Because many studies have only analyzed the tumor mass present within bone, it is unclear whether the redirection of tumor growth from intra- to extra-osseous sites is a widespread phenomenon. Taken together, the evidence suggests that bisphosphonate-induced prevention of bone disease is insufficient to prevent tumor progression, which eventually occurs even following very intensive therapeutic scheduling.

Effects on Extraskeletal Tumors

Whereas many studies have convincingly shown that bisphosphonates limit the extent of lytic bone disease, evidence for their direct antitumor effects is limited. High doses of bisphosphonates have been shown to significantly reduce growth of human tumor cells implanted subcutaneously in immunocompromised mice.^106,107 In contrast, when clinically achievable doses and incubation periods are used, there is generally no effect on tumors at peripheral sites. One notable exception is a report showing that in a lung cancer model, a dose of 1 μg/kg zoledronic acid given once weekly for 3 weeks reduced the growth of subcutaneously implanted tumors and increased survival.¹⁰⁸ These data need to be independently confirmed. The reports published to date have shown that doses of bisphosphonates exceeding those in clinical use are required in order to affect extraskeletal tumor growth.

Prevention Versus Treatment

As noted above, bisphosphonates are currently used to treat patients with evidence of bone metastases; whether they also act to reduce the spread of tumor cells to the skeleton remains to be established. A number of different in vivo studies have explored the effects of preventive protocols on subsequent tumor growth. In these models, animals received either bisphosphonates or placebo a few days prior to inoculation of tumor cells, and the number and extent of tumors were compared between the two groups. By the time tumor cells arrive in bone, the bisphosphonates will have already inhibited osteoclast activity, potentially reducing tumor cell adhesion to bone and lowering the levels of bone-derived growth factors in the bone microenvironment. Taken together, the data from preventive studies in a number of different tumor types (including breast and prostate) show that by administering bisphosphonates prior to tumor cell injection, there is a reduction in the level of bone disease (number and size of lytic lesions) accompanied by reduced tumor burden and lower incidence of metastases compared to control.^109,110 The preclinical data strongly suggest that bisphosphonates may have additional effects on tumor growth if therapy is initiated at early stages of disease, as opposed to only in the advanced (metastatic) setting. Several recent studies suggest a decreased incidence of prostate and breast cancer in patients receiving adjuvant bisphosphonate therapy.^111-113 Going forward, the clinical challenge will be to select which patients are at highest risk of developing future bone metastases and therefore most likely to benefit from adjuvant bisphosphonates.

Promise in Combination Therapy

There has recently been a shift in the focus of cancer research from the analysis of only tumor cells to also consider the contribution of the host microenvironment to tumor growth. We now understand a great deal more about how tumor cells interact with the surrounding normal cells, and how important these interactions are for the many steps involved in disease progression. Single-agent therapy has proven unsuccessful in eliminating tumors, and combination therapy with drugs that target both the tumor cells directly (chemotherapy, biological agents) and agents that modify their microenvironment (antiangiogenic and antiresorptive agents) is becoming an active area of focus. Bisphosphonates may hold their greatest promise as antitumor agents when used in combination with cytotoxic drugs. Several in vivo studies have reported substantially increased tumor inhibition and improved survival when bisphosphonates are added to standard chemotherapy regimens.^114-117 This has been extensively explored in models of both early and advanced breast cancer. A 6-week course of sequential administration of doxorubicin, followed 24 hours later by zoledronic acid, was shown to cause nearly complete elimination of subcutaneously implanted tumors.¹¹⁸ A subsequent study revealed that this anti-tumor effect was sustained for many months following cessation of treatment, supporting inclusion of zoledronic acid as part of a combination regimen in breast cancer.¹¹⁹ In both studies there was no significant reduction in tumor burden in animals receiving zoledronic acid alone. Similar data were reported from models of breast cancer–induced bone disease, where only animals receiving sequential doxorubicin and zoledronic acid had a reduction in intra-osseous tumor burden.¹²⁰ Although zoledronic acid did reduce the extent of cancer-induced bone disease in this model, tumor burden was unaffected, suggesting that simply eliminating bone disease is insufficient to prevent further tumor growth. A recent report of the effects of risedronate and docetaxel in a prostate cancer model has shown similar results, where the highest degree of inhibition of bone metastases was caused by a combination of the two agents.¹¹⁶ As was reported for zoledronic acid, the effect of risedronate alone was limited to preventing cancer-induced bone disease. There is now extensive evidence from in vivo models supporting the use of bisphosphonates in combination with cytotoxic therapies, and clinical studies are urgently needed to establish whether these promising results can translate to improved outcome for cancer patients.

RANK Ligand (RANKL) and OPG

An attractive target for blocking metastatic activity is the RANKL signaling triad that plays a dominant role in osteoclastogenesis: RANK, RANKL, and OPG. As noted above, denosumab, a human monoclonal antibody that specifically binds RANKL, is in phase III development.

RANKL, when bound to RANK on the surface of osteoclast precursors, promotes osteoclastogenesis. Conversely, OPG, a soluble decoy receptor for RANKL, inhibits osteoclast formation. An excessive production of RANKL by osteoblasts plays a key role in the pathogenesis of tumor-induced osteolysis, and inhibition of RANKL in animal models prevents tumor-induced bone destruction.

OPG and RANKL are both expressed by osteoblasts and bone marrow stromal cells, and the RANKL/OPG ratio determines the balance between new bone formation and resorption in healthy bone. In cancer, the RANKL/OPG ratio is perturbed by signals from cancer cells, causing excessive bone resorption or formation. PTHrP, IL-6, and IL-11 are secreted from cancer cells, increasing RANKL expression in osteoblasts.^121-124 IL-8 increases osteoclastogenesis in both a RANKL-dependent and -independent manner.¹²⁵ Recent data from osteoarthritis models demonstrate increased RANKL expression in osteoblasts treated with IL-1β, TNF-α, PGE2, and IL-17, which could also potentially play a role in cancer and bone interactions.¹²⁶ In addition to enhancing bone resorption in response to cancer cell signals, RANKL, RANK, and OPG may also function in cancer cell proliferation, migration to bone, and invasion of bone. Thus, targeting RANKL for bone metastases treatment may have antitumor effects that are both osteoclast-dependent and -independent.

Accumulating evidence indicates that cancer cells express all components of the RANKL signaling pathway, though the pathway function may be different in different types of cancers. The relative role of this pathway in cancer cells and its relationship to bone cells is under intense investigation. Numerous in vivo preclinical studies demonstrate the efficacy of targeting RANKL to treat bone disease associated with breast, prostate, lung, and colon cancer as well as melanoma, myeloma, and osteosarcoma. These effects appear to be osteoclast-dependent because reductions in osteoclast number are associated with reductions in tumor burden. However, recent data suggest that the RANKL pathway may also be important in skeletal complications in cancer that do not involve osteoclasts.

Homing of Cancer Cells to Bone

RANKL is clearly important in bone resorption and fueling cancer growth. However, some data suggest that bone-derived RANKL may also serve as "soil" or a chemoattractant to bone for RANK-expressing cancer cells, independent of bone resorption and osteoclast activity. In normal mammary epithelium, melanoma, and breast cancer cells, RANKL induced invasion and migration in vitro. These effects were blocked by OPG and were not observed in a RANK-negative colon cancer cell line. In a melanoma metastasis model that was previously shown not to activate osteoclasts and bone resorption, OPG did decrease bone metastases.⁴ Because OPG was effective in a system thought to be devoid of osteoclast activation, the authors concluded that the decrease in bone metastases was caused by inhibition of RANKL-mediated homing of cancer cells to bone and not by increased bone turnover. This conclusion is controversial as this mouse melanoma model may have some degree of osteoclast activation and increased bone turnover that was not recognized by the authors. However, recent unpublished data suggest that RANKL may have a role in tumor cell migration in other models.

Epithelial-to-Mesenchymal Transition (EMT) and Invasion

EMT and invasion are important determinants of cancer growth, as well as the establishment and subsequent proliferation of cancer cells within normal tissue. Functional RANKL expression by prostate cancer cells is correlated with EMT and a threefold increase in bone metastases,¹²⁷ supporting another tumor-specific role for RANKL signaling in cancer progression. In PC-3 prostate cancer cells, treatment with RANKL enhanced invasion of collagen matrix and increased expression of MMP-9 and IL-6; these effects were blocked with OPG.¹²⁸ Consistent with these results, RANKL was found to induce invasion of breast cancer cells in a matrigel invasion assay system.¹²⁹ The role of RANKL as a chemoattractant or transformant independent of osteoclasts has yet to be established in vivo, but in vitro studies are suggestive.

Apoptosis

Components of the RANKL signaling pathway may have additional effects on tumor cells. There is evidence that blocking RANKL with OPG could actually stimulate tumor growth by binding the pro-apoptotic factor TRAIL. Expression of OPG by MDA-MB-231 cells increased tumor inhibition of TRAIL and improved tumor survival.¹³⁰ OPG overexpression in MCF-7 breast cancer cells also enhanced tumor growth in bone,¹³¹ and OPG expression by breast cancer cells correlated with bone homing and colonization potential.¹³² Recent data also suggest that in MDA-MB-231 cells interaction between TRAIL and OPG could increase RANKL expression.¹³³ These data are in contrast to the inhibition of breast cancer growth in bone by systemic treatment with OPG in mice,¹³⁵ and may be due to the anti-apoptotic or perhaps autocrine effects of OPG when produced directly by tumor cells. Although there are concerns that systemic OPG therapy could enhance tumor cell survival, OPG treatment consistently decreases bone metastasis in murine models of breast cancer metastasis.

Angiogenesis

OPG expression in endothelium of malignant breast tumors correlated with higher tumor grade and appears to have proangiogenic effects in vitro. OPG was expressed in the endothelium of 59% of malignant breast cancers, but not in endothelium of nonmalignant tissue. OPG also supported endothelial cell survival in vitro and promoted cord structures in a matrigel tubule formation assay.¹³⁶ OPG may decrease apoptosis of both tumor cells and endothelium, simultaneously increasing tumor growth and necessary blood supply.

Other Bone-Targeted Therapies in Development

Other bone-targeted agents in earlier development inhibit cathepsin K, c-Src, TGF-β signaling, DKK1, endothelin A receptor, and other pathways involved in bone growth and resorption. Each of these agents has the potential to have antitumor effects that may be both bone cell-dependent and -independent. These agents are in various stages of clinical trials, and future studies will determine if true bone cell antitumor properties can be attributed to these agents.

< < PREVIOUS

NEXT > >