Over 80% of patients with advanced prostate cancer are affected by bone metastasis, an incurable progression of the disease that accounts for a vast majority of disease-related mortality and is associated with significant morbidity. In addition, androgen deprivation therapy (ADT), the mainstay of treatment for advanced prostate cancer, increases fracture risk and may contribute to skeletal morbidity. The combination of these factors makes bone metastases one of the most wide-reaching and difficult processes to address in patients with prostate cancer.
Because the precise molecular and cellular mechanisms of the metastatic process in prostate cancer are still poorly understood, there remains an unmet need for effective strategies to prevent the progression of disease or, at the very least, to diminish the likelihood for development of additional metastases. Among the various, mostly palliative, treatments used to treat bone metastases, bone-targeted approaches using bisphosphonates, radiopharmaceuticals, or endothelin receptor antagonists currently appear to offer the most promise in terms of efficacy and tolerability.
This chapter of the Report to the Nation on Prostate Cancer will review the best-understood mechanisms behind the development of bone metastases and the latest research in the use of agents to combat the progression of and the complications associated with bone metastatic prostate cancer.
Mechanisms of Bone Metastasis
Bone metastases in patients with breast cancer and multiple myeloma are predominantly osteolytic in nature. By contrast, prostate bone lesions are mixed, containing both osteolytic and osteoblastic elements. Under normal conditions, the rates of bone formation and bone resorption are balanced. In patients with prostate cancer, however, the increase in osteoblast number and activity, coupled with an increase in osteoclastic activity adjacent to the osteoblastic lesions,[10,11] leads to an imbalance in remodeling and to more brittle bone tissue. Of note, urinary deoxypyridinoline, a marker of osteoclastic activity, has been shown to predict skeletal complications in patients with androgen-independent bone metastatic disease,[ indicating that bone resorption can directly affect clinical outcomes.
Transforming growth factor-beta (TGF-beta), epidermal growth factor (EGF), and other bone-derived growth factors that promote prostate cancer cell growth and differentiation have been shown to influence the preferential localization of prostate cancer in bone tissue. TGF-beta, which is expressed by prostate cancer cells and plays a role in osteoclastic maturation, facilitates cellular adhesion to the bone matrix, which, in turn, influences expression of androgen-independent cellular growth. Similarly, by promoting migration of prostate cancer cells to bone tissue, EGF facilitates development of metastases in bone.
Other tumor-derived factors that have been shown to influence development of bone metastases include endothelin-1, osteoprotegerin, bone morphogenic protein, and insulin-like growth factor. Discussion of their respective roles in the pathogenesis of bone metastases and of their potential utility as therapeutic targets is presented below.
The primary mechanisms of bone metastasis development in prostate cancer are illustrated in the Figure.
Figure. Interaction between tumor cells and bone in patients with bone metastatic prostate cancer.
TGFβ = transforming growth factor-beta; IGF = insulin-like growth factor; ET-1 = endothelin-1; BMPs = bone morphogenic proteins; OPG = osteoprotegerin; EGF = epidermal growth factor; bFGF = basic fibroblast growth factor.
Detection of Prostate Bone Metastases
Bone scintigraphy, or conventional bone scan, with Tc-99m MDP radiotracer is the current standard of care for detection of bone metastases. With this method, the entire skeleton is imaged, and uptake of the tracer is mediated by osteoblastic activity rather than the tumor itself. Bone scans are very sensitive, but not specific; other processes are associated with increased tracer accumulation in bone, including trauma, infection, and degenerative joint disease such as arthritis.
Studies with magnetic resonance imaging (MRI) indicate improved efficacy over bone scintigraphy due to its ability to detect metastatic cellular foci (ie, fatty and cellular tissue elements of bone marrow) before cortical bone destruction has occurred. In a series of 36 prostate cancer patients, MRI showed to be particularly useful in clarifying inconclusive radiographic tests in prostate cancer. Of 19 positive bone scans in 36 patients, MRI confirmed 8 equivocal cases and demonstrated additional metastatic lesions in 6 patients. MRI also indicated areas of spinal compression in 5 patients with spinal metastases, and altered the clinical stage in 2 patients. Only 1 false positive was noted. Results from a second trial in 19 men with prostate cancer concurred: Detection of metastatic disease was seen in 7% (1/13) of patients with a negative bone scan, and in 2 of 4 patients with indeterminate bone scans. Again, only 1 false positive was reported.
Given its demonstrable sensitivity over bone scintigraphy, MRI might also play a role in initial staging. In a retrospective review of staging skeletal scintigraphs in 200 patients with breast or prostate cancer, researchers determined that 7 of 100 breast cancer patients and 9 of 100 prostate cancer patients would be expected to have a negative MRI and abnormal bone scintigraphy. However, in the breast cancer patients, 1 of 7 abnormal scans represented metastatic disease, while in prostate cancer patients, only 1 of 3 represented metastatic disease. In addition, the high cost and difficulty in scanning all areas of interest limit the routine use of MRI. Thus, although data suggest that the addition of MRI to scintigraphy in initial staging may provide a more accurate basis for patient management, further study of MRI in this setting is clearly needed.
Because of its ability to detect tumors on the basis of metabolic activity rather than by simply demonstrating increased bone mineralization, the use of fluorine-18-deoxyglucose positron emission tomography (FDG-PET) has been examined as an alternative to bone scintigraphy. However, although it has proven effective in detecting other solid tumors, results in prostate cancer have been disappointing, possibly because prostate cancer tends to be metabolically quiescent. In one study, the technique underestimated the extent of osseous metastatic lesions in 4 of 5 patients, while in another it missed many of the osseous lesions found by bone scans.
By contrast, the use of 18-fluoride (18-F) as a tracer has shown greater promise. Early studies showed it to be more sensitive than bone scintigraphy, detecting twice as many prostate metastases in patients with established disease. Upon evaluation of patients with known metastatic lung cancer, two studies demonstrated that 18-F PET was more accurate and resulted in fewer false negatives compared with both bone scintigraphy and single photon emission computed tomography (SPECT). In a study of 26 patients with solid tumors and bone metastases, the integration of PET and computed tomography (CT) using the 18-F tracer had a 100% sensitivity and an 88% specificity vs 88% and 56%, respectively, with PET alone. Because these techniques are more expensive than the standard bone scintigraphy, additional studies with larger patient populations will be needed to confirm the benefits of 18-F PET and/or 18-F PET/CT.
Management of Bone Metastases
Bisphosphonates, synthetic analogues of inorganic phosphate that affect human bone metabolism through direct and indirect actions on osteolysis, have emerged as a key therapeutic strategy in the management of bone disease. Although they have demonstrable efficacy in treating osteoporosis, Paget's disease of bone, and hypercalcemia of malignancy, and have also proven beneficial in treating skeletal complications and pain in breast cancer, their application in the management of advanced prostate cancer is expanding, with more research being conducted to fully understand the potential of these agents.
Bisphosphonates are believed to act through several mechanisms: inhibition of cancer cells that bind to the bone matrix; osteoclast apoptosis through competitive inhibition of ATP/ADP translocase; inhibition of osteoclast formation, migration, and bone resorption; and inhibition of matrix metalloproteinases (MMPs). Recent in vitro data also suggest that bisphosphonates might act directly on osteoblasts by promoting preosteoblastic cell growth and differentiation. Their use in stemming bone loss seen with ADT has been well described, and is discussed in a separate chapter.
Variations in the molecular structure of each bisphosphonate determine its affinity for the bone mineral surface and its relative potency. Because the ability of bisphosphonates to inhibit cancer cell adhesion to cortical and trabecular bone has been shown to predict clinical efficacy, it has been suggested that use of more potent agents might improve outcomes in the management of bone metastases.
Of the available bisphosphonates, the third-generation zoledronic acid is the most potent. Like other agents in this class, its effects are mediated primarily through osteoclast proliferation and apoptosis.
In one of the largest phase 3 clinical trials with zoledronic acid, 643 patients with asymptomatic or minimally symptomatic bone metastatic androgen-independent prostate cancer (AIPC) were randomized to placebo or to 4 mg or 8 mg of zoledronic acid once every 3 weeks. To minimize renal complications, the protocol was amended to administration of the infusion over 15 minutes rather than over 5 minutes, and the infusate volume was increased from 50 to 100 mL. In addition, patients in the 8-mg arm were switched to the 4-mg dosage.
At study end, after 15 months of follow-up, fewer patients enrolled on the 4-mg zoledronic acid arm vs the placebo arm experienced at least one skeletal complication (33.2% with zoledronic acid vs 44.2% with placebo; P = .021), and the time to the first occurrence of any skeletal complication was shorter with placebo (median time to first event not reached in 4-mg group with zoledronic acid vs 321 days with placebo; P = .011). The skeletal morbidity rate, calculated as the number of skeletal complications divided by the time at risk in years, was significantly lower with 4-mg zoledronic acid (.80 vs 1.49; P = .006). Significant reductions in biochemical markers of bone metabolism were also observed. The drug was well tolerated; the most commonly reported adverse events included fatigue, anemia, myalgia, fever, and lower-limb edema.
Updated results from this trial demonstrated continuing benefits at 24 months. The proportion of patients experiencing at least one skeletal complication remained significantly lower with 4-mg zoledronic acid over placebo (38% vs 49%; P = .028), as did the median time to first event (488 days vs 321 days; P = .009). A significant delay was noted in the median time to a second event in patients on the zoledronic acid arm (median not reached for 4-mg zoledronic acid vs 449 days for placebo; P = .006), and a 32% relative reduction was noted in the percentage of patients on the zoledronic acid arm experiencing a second event vs those on placebo (P = .017). Notably, a multiple-event analysis combining skeletal complication rate and timing suggested an overall 40% reduction in the risk for developing subsequent events with 4-mg zoledronic acid vs placebo (P = .011).
As a class, bisphosphonates have an analgesic effect on bone pain, which is often included as an endpoint in clinical studies on bone metastases. In this trial, although there was an increase in mean bone pain scores (as measured by the Bone Pain Inventory) from baseline in both treatment and control groups, bone pain was consistently lower in patients treated with zoledronic acid vs placebo at all time periods, reaching statistical significance at the 3-, 9-, 21-, and 24-month time points.
On the basis of results from this and other studies, zoledronic acid was approved for the management of bone metastases in patients with AIPC, and may be of particular value for patients at high risk for bone fractures or spinal cord compression. The effect of zoledronic acid on skeletal complications in patients with hormone-sensitive disease will be evaluated in the CALGB/CTSU 90202 trial that is currently recruiting patients. Additional studies are needed to assess the potential for bisphosphonates to prevent the development of bone metastases.
Pamidronate and Clodronate
Pamidronate disodium, a second-generation bisphosphonate, and clodronate, a first-generation bisphosphonate, are far less potent than zoledronic acid, and the evidence supporting the use of either agent in prostate bone metastases is less clear than that for zoledronic acid.
Pamidronate is approved in the United States for use in hypercalcemia of malignancy, as well as for the treatment of osteolytic bone metastases in patients with breast cancer and multiple myeloma. However, in patients with symptomatic bone metastatic AIPC, data from two trials pooled for analysis showed no significant benefit for pamidronate 90 mg every 3 weeks vs placebo. Of the 374 patients enrolled in the trials, no sustainable differences were noted between the two arms in the proportion of patients with a skeletal complication or in change from baseline in mobility at 9 weeks or at 27 weeks. Of the 301 patients assessable for pain at 9 weeks and 218 patients assessable at 27 weeks, Brief Pain Inventory scores declined only minimally from baseline and were comparable between the treatment and control groups. The only significant difference was noted in a subset of pamidronate patients with decreasing or stable analgesic use, in whom the mean decrease in pain scores was significantly greater than placebo at 9 weeks (P = .008 for worst pain; P = .011 for average pain), but not at 27 weeks.
Data with clodronate, which remains investigational in the United States, are even less promising. Of the 209 patients with symptomatic bone metastatic AIPC receiving mitoxantrone and prednisone, no significant differences were noted between the addition of clodronate 1500 mg every 3 weeks or placebo in the primary study endpoint of palliation, defined as a reduction of 2 points on the pain scale or a 50% reduction in analgesic intake. No differences were seen in other endpoints of median duration of response, progression-free survival, overall survival, and overall quality of life, although patients with more severe pain fared slightly better on subgroup analysis.
In patients with androgen-sensitive bone metastatic prostate cancer, investigators attempted to increase the dose of oral clodronate to more closely match that which would have been delivered intravenously. However, although the 2080-mg daily dose of clodronate resulted in a 21% reduction in the risk of symptomatic bone progression or prostate cancer death and a 20% reduction in the risk of death, clodronate treatment was associated with significantly greater risk for gastrointestinal problems, leading to dose modification in approximately one third of patients (hazard ratio, 181% increase; P < .0001). The high rate of gastrointestinal toxicities suggests that the increased dose strategy is not feasible, and that the use of clodronate and other oral bisphosphonates in patients with prostate cancer is unlikely to be effective in management of bone metastases.
Although subgroup analysis in this study demonstrated a significant decrease in the time to first symptomatic skeletal complication, suggesting that there might be a benefit toward initiating bisphosphonate therapy at an earlier time point in the management of bone metastatic AIPC. Results from a similarly designed study ultimately found no significant benefit with clodronate in preventing symptomatic bone metastases or in improving overall survival. Clinical trials with agents more potent than clodronate might be better able to address this hypothesis.
For patients with symptomatic bone metastases in whom bisphosphonates confer only moderate effects, radiation-based solutions can provide some palliative benefit. The use of local radiation therapy in this population as a single modality is well documented, and has been shown to provide symptomatic relief within a few weeks of administration. However, this approach is limited to patients with few sites of disease; hemibody radiotherapy, which targets the lower half of the body with a single dose of radiation, has been used in patients with more widespread disease. Both of these approaches have demonstrated significant improvements in pain scores and significant delays in the development of new sites of pain. Nevertheless, their lack of specificity for tumor bone sites can damage normal tissue, and the severe gastrointestinal toxicity seen with hemibody radiotherapy make these treatment modalities less than optimal.
By contrast, radioisotopes preferentially accumulate in tumor bone sites, minimizing the local tissue damage caused by standard radiotherapy. Studies with strontium-89, the most widely used radioisotope, demonstrated equivalence to local radiotherapy or hemibody radiotherapy in pain relief and in progression-free survival and time to disease progression. Response rates are above 70%, although about 20% experience pain flares 1-2 weeks after therapy. When added to consolidation therapy with doxorubicin, strontium-89 also showed activity, significantly increasing median survival time from 16.8 months to 27.7 months. However, hematologic toxicity with strontium-89, mainly transient dose-dependent thrombocytopenia, is frequent, with reports of patients developing acute myeloid leukemia following therapy.
A recent phase 3 trial of samarium-153 also demonstrated positive results, but with fewer side effects. By study end, significant improvements in analgesic consumption and pain were seen with the radioisotope vs placebo; changes in visual analog pain scores correlated with opiate use in the treatment arm but not in the placebo arm. Of note, hematologic toxicity was mainly mild transient myelosuppression, with nadirs observed 3-4 weeks after initiation of therapy and with recovery to normal levels observed by week 8.
Research with rhenium-188-HEDP, a radioisotope with a shorter physical half-life than strontium-89, is promising. In a phase 2 study in patients with hormone-refractory metastatic disease, palliation was achieved in 60% of patients receiving one dose of the isotope and in 92% receiving two doses; no grade 3 or grade 4 hematologic toxicities were seen in either group.
Continued research with these agents will explore their role in preventing bone metastasis and slowing disease progression; further attempts to enhance the efficacy of radiation-based therapies in combination with bisphosphonates and chemotherapy are also being explored.
Despite therapy, the prognosis for men with prostate cancer who develop bone metastases is poor. On average, median survival is currently estimated to be 30-35 months. Hence, therapeutic strategies that move beyond palliation to prevention of disease progression and development of bone metastases are essential, both in terms of prolonging life and in improving the prospects for better quality of life with fewer skeletal complications.
Although practicing clinicians have some ability to determine which patients are at risk for the development and progression of bone metastatic prostate cancer, the ability to extrapolate information about cellular or molecular mechanisms that potentially contribute to the metastatic cascade to assist with management is not as clear-cut.
Fortunately, researchers continue to define novel targets through which the skeletal metastatic process might be inhibited or interrupted. Some of the most compelling advances in this regard have been bone-targeting strategies using bisphosphonates to reestablish bone homeostasis, and ET-A antagonists to delay tumor progression in bone. Nevertheless, continued research in this arena is of paramount importance, and future clinical trials directed at novel targets such as RANKL, MMP, and OPG, which are currently in experimental development, are eagerly awaited.
These and other key issues regarding patients with bone metastatic prostate cancer are summarized in Table 1.
Optimal management of bone metastatic prostate cancer ultimately relies on discovering a means by which to halt disease progression altogether. As described in Table 2, future research and improvements in clinical practice in this and other areas will greatly contribute to our understanding of bone metastatic prostate cancer. In the interim, the challenge remains to diagnose bone metastatic prostate cancer at its earliest stages and as precisely as possible, so that skeletal complications can be adequately addressed and overall quality of life can be improved.
Table 1. Management of Bone Metastases in Patients With Prostate Cancer: Summary
1 The cross-talk among tumor cells, osteoclasts, and osteoblasts in the development of bone metastatic prostate cancer results in an imbalance in remodeling and subsequent fragile skeletal bone tissue as well as increased tumor growth in cells that have colonized bone tissue. 2 BMPs and RANK/RANKL mediate osteogenic activity, thereby making them attractive targets for disruption of bone disease progression. 3 Bone scintigraphy with Tc-99m tracer remains the standard of care for detection of bone metastases. Preliminary results with MRI, 18-F PET, and 18-F PET/CT have shown promise in improving sensitivity and specificity. 4 Administration of radiopharmaceuticals, such as strontium-89, samarium-153, and rhenium-188-HEDP, has demonstrated improvements in palliation and less damage to normal tissue compared with local radiation or hemibody radiation. 5 The potent bisphosphonate zoledronic acid has demonstrated an ability to reduce skeletal complications. Ongoing studies will address the role of earlier therapy.
Table 2. Future Research and Practice Opportunities: Calls to Action
1 Develop practice guidelines for the use of bisphosphonates in men initiating androgen deprivation therapy to prevent disruptions in bone metabolism and slow progression to bone metastatic disease. 2 Promote early diagnosis of bone metastatic disease by increased screening and thorough evaluations of patients with prostate cancer. 3 Encourage study of combination therapy with investigational and approved bone-targeted agents to stem the progression of bone metastases. 4 Develop criteria for conducting clinical trials in patients with bone metastatic disease with cytostatic agents, including clinically meaningful endpoints. 5 Promote enrollment in clinical trials by urologists, radiation oncologists, medical oncologists, and patients at all stages of prostate canc