The clinical application of BTAs in current clinical practice is mostly focused on the inhibition of osteoclast activity either by the use of BPs or denosumab. The radiopharmaceutical agent radium-223 (Ra-223) is also considered by many authors as a BTA. Indeed, it has high affinity to bone, particularly to the site of bone metastasis, and does reduce significantly the incidence of bone complications in patients with prostate cancer and bone metastases. However, Ra-223, which is an alpha emitter, is not just a BTA because it also causes direct cancer cell death by radiation-induced double strand breaks. In our perspective, Ra-223 is a mixed therapeutic proposal (BTA and anti-cancer agent). We do not know yet whether Ra-223 can also be effective in targeting osteoblast hyperactivity (like in blastic bone metastases associated with prostate cancer).
We shall describe first the major achievements of BTAs agents in the treatment of bone metastases followed by a brief exposition of their role in the adjuvant setting.
3.1. Bone Targeted Agents in Advanced Disease
Solid tumors with bone metastases are usually complicated by the occurrence of skeletal related events (SREs), a composite endpoint frequently defined in clinical trials by the occurrence of pathologic fractures (incidental or symptomatic), spinal cord compression or radiation/surgery to the bone; in some, hypercalcemia of malignancy is also considered. Symptomatic skeletal events (SSEs) differ from the latter for only including symptomatic pathologic fractures.
SREs are a frequent clinical event in patients with bone metastases and inflict a significant burden to cancer patients. In prostate cancer, bone metastases occur in up to 70% of cases and, over a two-year period, SREs occur close to half of patients with metastatic castration resistant prostate cancer (mCRPC) with bone metastasis not treated with a BTA [68
]. Bone is also the site of first disease recurrence in 30% to 40% of women with breast cancer and approximately 70%–80% will develop bone metastasis throughout the course of the disease, with the complications associated causing significant morbidity and impairing quality of life. In these patients, pain is the most frequent SRE, occurring in up to 75% of patients, pathologic fracture follows with 16%, while the least common is spinal cord compression in 3% [69
]. Although more frequent in breast and prostate cancer, bone metastasis can occur in virtually any type of cancer.
When assessing the radiographic appearance of bone metastases, three patterns occur: osteoblastic, osteolytic or mixed, as a function of radiographic density; however, all are associated with an increase in osteoclastic activity, including the osteoblastic metastasis, so multiple osteoclast targeted agents have been studied in this setting [70
]. In current clinical practice, only denosumab and BPs are approved for the prevention of SREs in metastatic disease (Table 1
). As detailed above, Ra-223 is also considered by some as a BTA.
BPs have high affinity for calcium ions thus attaching to hydroxyapatite binding sites on bone surface, especially those undergoing active resorption. During bone resorption, BPs are internalized by bone-resorbing osteoclasts and inhibit osteoclast function [71
]. Nitrogen-containing BPs (alkyl-amino BPs: pamidronate, alendronate, ibandronate; heterocyclic BPs: risendronate, ZA) impair the mevalonate pathway by inhibiting the farnesyl diphosphate synthase (FPP synthase), ultimately preventing prenylation of small GTPase signaling proteins vital for normal cellular function. Non-nitrogen containing BPs (etidronate, clodronate, tiludronate) induce to the formation of deleterious metabolites in osteoclasts. Beyond their effects on osteoclast inhibition, BPs may also have antitumor and/or antiangiogenic effects, but this is a controversial area (see Section 3.4
.). Investigations are ongoing to better define the clinically relevant effects of BPs in patients with cancer [72
]. Pamidronate and ZA have been approved by both European Medical Agency (EMA) (or local European authorities) and Food and Drug Administration (FDA) for the treatment of skeletal metastases from solid tumors and multiple myeloma (MM). Clodronate is not approved for clinical use in the US but is available in Europe. Ibandronate is also an alternative. ZA is the only BP approved for mCRPC and is also approved for use in patients with other solid tumors.
The first positive clinical trial using BPs in breast cancer goes back to 1987, where 34 normocalcemic breast cancer patients with progressive osteolytic bone metastases were treated with clodronate (1.6 g/day) or placebo for 12 months. Bone pain, extension of bone metastases and formation of new osteolytic foci were reduced by clodronate, and development of severe hypercalcemia was prevented [81
]. This was followed by a larger, double-blinded trial in 1993, where 173 patients with bone metastasis due to breast cancer were randomized to oral clodronate (1.6 g/day) or placebo. The proportion of all SREs was substantially reduced (218.6 vs. 304.8 per 100 patient-years; p
< 0.001). A trend in favor of clodronate was also found in the rates of non-vertebral fracture and need of radiotherapy for bone pain control (particularly spinal pain) [82
Pamidronate, a second generation and oral BP, was investigated on a subsequent randomized study aiming to assess its effects on the morbidity from bone metastases of breast cancer patients and its gastrointestinal tolerability. In the pamidronate group, the occurrence of hypercalcemia of malignancy, symptomatic impending fractures, need for radiotherapy and severe bone pain improved by 65%, 50%, 35% and 30%, respectively [83
Pamidronate was also evaluated in two double-blind, randomized placebo-controlled trials at a dose of 90 mg IV in breast cancer patients who were being treated either with chemotherapy or hormonal therapy. In this study, pamidronate treatment was associated with a significant reductions in SREs and pain. In patients receiving chemotherapy, median time to first SRE was longer in the pamidronate group when compared to placebo (13.1 vs. 7.0 months, p
= 0.005) and lower proportion of patients with any SRE (43% vs. 56%, p
= 0.008). In patients receiving hormonal therapy, at 24 cycles, the proportion of patients with an SRE was 56% in the pamidronate group and 67% in those treated with placebo (p
= 0.027); median time to first SRE was 10 vs. 7 months (p
= 0.05) for the pamidronate and placebo groups, respectively [84
ZA, a third generation BP, was studied in comparison to pamidronate. In a pivotal double-blind randomized trial of patients with breast cancer and bone metastases, researchers compared ZA at 4 or 8 mg IV to pamidronate at 90 mg IV with no significant differences emerging between these agents in terms of the number of SREs or time to first SRE [86
]. ZA is also the only BP approved for the prevention of SREs in mCRPC. Clodronate or pamidronate were not effective in improving the number of SREs in these patients [87
]. In the ZA 039 trial of mCRPC, 643 patients were randomly assigned to the treatment with ZA at 8 or 4 mg IV or placebo q4w showing a significant reduction in the rate of SREs (49% vs. 38%, p
= 0.0029) and an increase in the median time to first SRE in favor of ZA at 4 mg (the 8 mg arm was discontinued and patients reassigned to the 4 mg due to renal toxicity) [68
]. ZA at a 4 mg IV dose q4w was also studied in the CALGB 90202 trial, in the castration-sensitive setting for advanced prostate cancer with bone metastasis. The study was terminated prematurely (645 of 680 planned accruals), because of the withdrawal of sponsor support. The median time to first SRE was 31.9 months in the ZA group vs. 29.8 with placebo (hazard ratio (HR) 0.97; p
= 0.39), therefore not supporting the use of ZA in this setting [92
Finally, ZA was further compared with ibandronate. In the ZICE trial, a non-inferiority study, ibandronate at the dose of 50 mg/daily was compared to ZA at the dose of 4 mg every 3 to 4 weeks in patients with breast cancer and bone metastasis [79
]. Ibandronate was not non-inferior, with an annual rate of SREs of 0.499 (95% CI 0.454 to 0.549) for ibandronate and 0.435 (0.393–0.480) for ZA. However, given its safety profile and the fact that it is the only oral available BP some consider it as an alternative agent for unfit patients for which frequent hospital visits are not feasible [93
Data from clinical trials with BPs in other solid tumors different than breast and prostate cancer are scarcer. ZA was evaluated in a placebo-controlled trial of 773 patients with skeletal metastases from cancers other than breast and prostate. Patients were randomly assigned to IV ZA at 8 mg, 4 mg or placebo q3w with concomitant antineoplastic therapy. As in other trials, ZA at 8 mg was subsequently reduced to 4 mg due to renal adverse events. SRE incidence was reduced in both ZA groups (38% for 4 mg and 35% for 8/4 mg of ZA vs. 44% for placebo; p
= 0.127 and p
= 0.023 for 4-mg and 8/4-mg groups, respectively). Moreover, ZA increased time to first SRE in the 4 mg group (median 230 days vs. 163 days for placebo; p
= 0.023) [77
]. Based upon these results, the use of ZA is recommended in patients with advanced solid tumors with evidence of bone metastases.
Bone disease is also a typical feature of multiple myeloma (MM). As a matter of fact, it is the cancer presenting most frequently bone metastasis (in up to 90% of patients). Bone lesions in MM are purely osteolytic (low radiographic density) with up to 60% of patients developing pathologic fractures over the course of their disease [95
]. The efficacy of IV pamidronate in this setting was demonstrated in a clinical trial. Patients with stage III MM and at least one lytic lesion received either placebo or pamidronate 90 mg IV. The proportion of patients who developed any SRE was lower in the pamidronate group (28% vs. 44% p
= 0.015) and the mean number of SREs per year was inferior in the pamidronate group (1.3) than in placebo-treated patients (2.2; p
= 0.008) [96
]. Similarly to breast cancer, ZA and pamidronate reduced SREs in a comparable fashion to those seen in MM patients [11
In addition to BPs, osteoclast inhibition can also be achieved by targeting RANKL with denosumab, a fully humanized monoclonal antibody that binds to RANK and has demonstrated superiority to ZA in patients with bone metastatic disease from breast and prostate cancer. Side to ZA, denosumab is considered by international medical associations as an alternative treatment for patients with solid tumors and bone metastases [97
]. Despite this equal indication; denosumab has demonstrated superiority to ZA in terms of time to first and subsequent SREs in patients with bone metastatic disease from breast and prostate cancer.
ZA was compared to denosumab for SRE reduction in three double-blind phase 3 studies, comparing ZA 4 mg IV q4w vs. denosumab 120 mg q4w. In Trial 20050103, which included 1904 patients with mCRPC, denosumab delayed the time to first SRE by 18% (20.7 vs. 17.1 months, HR 0.82, p
= 0.0002) [74
]. In Trial 20050136, 2046 patients with metastatic breast cancer involving the bone, denosumab delayed the time to first SRE compared to ZA (32.4 vs. 26.4 months HR 0.82, p
= 0.01) [73
]. Denosumab has also been studied in a variety of other malignancies in Trial 20050244 that compared denosumab and ZA in 1776 patients with MM or bone metastases from a solid tumor other than breast or prostate cancer. In this study denosumab was non-inferior to ZA in delaying time to first SRE (20.6 vs. 16.3 months, HR 0.84; 95%, p
= 0.0007), but not statistically superior to ZA in delaying time to first SRE (p
= 0.06) [75
3.4. Adjuvant Use of Bone Targeted Agents
Several studies and a recent meta-analysis showed that BTAs are useful drugs to reduce the risk of breast cancer recurrence in postmenopausal women; its role in other tumors, namely prostate cancer, is less clear (Table 2
Beyond the antiresorptive effects of BPs that reduce SREs and treatment related bone loss, pre-clinical evidence further suggests an anti-cancer activity for BPs [101
]. In specific, BPs may act directly as cytotoxics, or indirectly by inhibiting angiogenesis, inhibiting the recruitment of tumor associated macrophages (TAMs), stimulating γδ T cells, or by acting synergistically with other chemotherapeutics. Finally, both BPs and denosumab disrupt the “vicious cycle” thus limiting tumor cells access to growth factors entrapped in the bone matrix. However, despite the strong preclinical rational for the anti-cancer action of BPs and denosumab, clinical studies performed in the metastatic setting never demonstrated a survival benefit. Furthermore, subsequent individual studies performed in the adjuvant setting intending to improve recurrence and survival and integrating BTAs in the conventional adjuvant regimen were inconsistent whether or not BTAs are useful drugs, especially for cancers other than breast. A recent meta-analysis of breast cancer studies added some clarity. Selected individual studies and the referred meta-analysis are detailed bellow, first for breast cancer and after for prostate cancer.
For breast cancer, the ABCSG-12 was a seminal phase III study showing a reduction in the risk of recurrence for patients treated with ZA [102
]. In this trial, 1803 premenopausal women with endocrine-responsive early breast cancer were treated with goserelin plus tamoxifen or aromatase inhibitors with or without ZA (4 mg every six months for three years). Those treated with ZA had a 23% reduction in the risk of disease recurrence (absolute reduction of 3.4%; hazard ratio (HR) 0.77 (95% CI 0.60–0.99), p
= 0.042) and a strong trend towards improved survival (HR 0.66 (95% CI 0.43–1.02), p
= 0.064). An even more pronounced effect was noted for patients 40 years or older.
The AZURE trial was a subsequent study intending to definitely solve this question. In this phase III randomized trial that included 3360 patients with axillary lymph-node metastasis or a T3–T4 primary tumor irrespective of menopausal status, after 59 months of median follow-up, those receiving ZA had an unimpressive 2% improvement in the relative risk of recurrence (DFS; HR 0.98 (95% CI 0.85–1.13), p = 0.79) and a non-significant 15% reduction in the risk of death (OS; HR 0.85 (95% CI 0.72–1.01), p = 0.07). Of note, in a planned sub-group analysis, late postmenopausal women (>5 years) had a remarkable benefit from ZA: a relative 25% reduction in the risk of recurrence (DFS) and 26% in the risk of death (OS). The fact that all patients in the ABCGS-12 were functionally postmenopausal (under ovarian suppression treatment) seems to reconcile the results of ABCSG-12 and AZURE trials.
More recently, a large Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) patient-level meta-analysis of trials of adjuvant BPs (amino and non-amino BPs) including 18,776 women with breast cancer showed a 18% reduction in the risk of death from breast cancer in the subset of postmenopausal women (HR 0.82 (95% CI 0.73–0.93), p
= 0.002) [9
]. This effect seems to mainly stem from the reduction of 28% in the risk bone recurrence (HR 0.72 (95% CI 0.60–0.86), p
= 0.0002) and not from extra-osseous recurrences, namely loco-regional or visceral. Admitting the usefulness of adjuvant BPs, SWOG0307 trial is comparing three different agents: ZA, clodronate and ibandronate. Preliminary results presented at the American Society of Clinical Oncology annual meeting 2015 suggest no difference between these agents in terms of DFS or grade 3/4 adverse events [104
]. However, a numerical difference in rate of osteonecrosis of the jaw was noted: highest for ZA (1.2%), then ibandronate (0.6%) followed by clodronate (0.3%). These data are consistent with the EBCTCG meta-analysis.
Denosumab is also being actively studied as an adjuvant treatment in breast cancer. Early data from the ABCSG-18 study, a phase III trial of 3425 postmenopausal patients with early hormone receptor positive BC receiving aromatase inhibitors treatment with or without denosumab (60 mg every six months), showed a reduction in the risk of recurrence of approximately 18% (HR 0.816 (95% CI 0.66–1.00), p
= 0.051) [105
]. The role of denosumab to improve cancer outcomes in the adjuvant setting (specifically bone related outcomes) is being prospectively tested in the D-CARE (NCT01077154) trial. Primary outcome is bone metastasis-free survival and estimated completion date of November 2017.
As for prostate cancer, adjuvant BTAs have a less established role. In the pivotal study ZEUS (Zometa European Study), a phase III trial that recruited 1433 patients with high-risk non-metastatic prostate cancer to be treated with or without every three months ZA, no difference was found in the proportion of patients developing bone metastases (17.1% vs. 17.0% without ZA; p
= 0.95) [107
]. In the RADAR study, a phase III trial recruiting 1071 men with locally advanced prostate cancer for the treatment with radiotherapy plus 6 or 18 month of androgen deprivation therapy (ADT) with or without 18 months of ZA, no differences were found in OS of prostate cancer specific survival between groups [108
]. However, those patients with high Gleason score (i.e., 8–10) and treated with ADT for 18 months derived a benefit from ZA (in the form of the reduction of PSA progression and need for secondary therapeutic intervention). Of concern was the increased proportion of patients with bone progression in the subgroup treated for six months with ADT plus ZA vs. ADT only. These subgroup analyses should be interpreted with caution. Longer follow-up and other trials might add some extra clarity at these results before definitive conclusions can be taken.
If the improved cancer outcomes discussed above derive from any direct or indirect anti-cancer action of BTAs or rather derive from the altered “soil” resulting from BTAs action at the bone microenvironment is still not established; however, early data from breast cancer pointing to a shared effect between BPs and denosumab supports the latter. Moreover, it is intriguing that up to now only breast cancer patients seem to benefit from this treatment in terms of cancer outcomes, and unexpectedly not all, but only the subgroup of postmenopausal women. The fact that in prostate cancer, bone metastases are often blastic reflecting a strong interaction between cancers cells and osteoblasts could be a possible explanation for the lack of efficacy of BTAs that behave as anti-osteoclasts agents only to prevent disease relapse in bone.