Effects of denosumab in patients with bone metastases with and without previous bisphosphonate exposure

Abstract

Bone metastases place patients at increased risk of skeletal-related events (SREs), including pathologic fractures, spinal cord compression, severe pain requiring radiotherapy or surgery, and hypercalcemia, because of increased osteoclast-mediated bone resorption. Denosumab, a fully human monoclonal antibody, decreases bone resorption by inhibiting RANKL, which mediates osteoclast activity. We compared the effects of denosumab in two phase 2 studies in patients with bone metastases naive to intravenous bisphosphonate therapy (IV BP; n = 255) and those with elevated levels of the bone resorption marker urinary N-telopeptide (uNTX) despite ongoing IV BP treatment (n = 111). Patients were randomized to receive IV BP every 4 weeks or subcutaneous denosumab every 4 weeks (30/120/180 mg) or every 12 weeks (60/180 mg). Patients treated with denosumab experienced a rapid and sustained reduction in bone turnover regardless of prior IV BP exposure. After 25 weeks, the median uNTX reduction was 75% (IV BP-naive) and 80% (prior IV BP) after denosumab treatment and 71% (IV BP-naive) and 56% (prior IV BP) in the IV BP arms. Denosumab patients with prior IV BP exposure had marked suppression of the osteoclast marker TRAP-5b (median reduction: denosumab 73%, IV BP 11%). SRE incidence was low across both studies. In patients previously treated with BPs, the rate of first on-study SRE was lower in the denosumab groups (8%) than the IV BP group (17%). Denosumab appeared to be well tolerated in both studies. Denosumab suppresses bone resorption markers independently of prior BP treatment, even in patients who appear to respond poorly to BPs. © 2010 American Society for Bone and Mineral Research.

Introduction

To maintain skeletal integrity and prevent skeletal complications such as pathologic fracture, severe bone pain requiring radiotherapy or surgery, spinal cord compression, and hypercalcemia in patients with bone metastases, treatment with intravenous bisphosphonates (IV BPs) is the current standard of care.1–3 In phase 3 clinical trials of patients with breast cancer, prostate cancer, and other solid tumors,4–6 38–43% of patients experienced a skeletal-related event (SRE) while receiving zoledronic acid, suggesting an unmet medical need and demonstrating the necessity for more effective therapy.

In bone metastases, a continuous cycle of tumor growth and osteolysis is marked by the activity of the receptor activator of NF-κB ligand (RANKL), which mediates the formation, function, and survival of osteoclasts.7–9 Denosumab is a fully human monoclonal antibody that specifically binds and neutralizes RANKL, inhibiting osteoclastogenesis and decreasing osteoclast-mediated bone destruction.10 Denosumab provides a potential option for the prevention of bone destruction caused by bone metastases or multiple myeloma or for the prevention and treatment of osteoporosis. A single subcutaneous (SC) dose of denosumab suppressed bone turnover for up to 6 months in postmenopausal women with low bone mass and for up to 12 weeks in patients with multiple myeloma or breast cancer,10, 11 including women with early-stage breast cancer receiving aromatase inhibitors.12

In this report, we compare the efficacy and safety of denosumab in reducing bone turnover in two clinical studies of patients with bone metastases associated with various solid tumors or multiple myeloma. In the first study, patients had no previous exposure to IV BPs. In the second study, patients had high levels of the bone turnover marker urinary N-telopeptide (uNTX) despite ongoing treatment with IV BPs. Evidence of bone turnover reduction is provided by decreased levels of bone turnover markers: uNTX and serum C-telopeptide (sCTX), which indicate bone resorption; tartrate-resistant acid phosphatase (TRAP-5b), a measure of viable osteoclasts that cause bone resorption; and bone-specific alkaline phosphatase (BSAP), procollagen-1 N-terminal peptide (P1NP), and osteocalcin, markers of bone formation. These bone turnover markers have been used widely in clinical studies and are recognized as reliable, accurate, and relevant for this therapeutic indication.13–15 We also report safety outcomes after up to 57 weeks.

Materials and Methods

Patients

Eligible patients were 18 years of age or older, with histologically confirmed cancers, radiographic evidence of at least one bone lesion, and an Eastern Cooperative Oncology Group (ECOG) performance status of 2 or less. Study 20040113 (NCT00091832) evaluated women with breast cancer and bone metastases who had not been treated previously with IV BPs.16 Study 20040114 (NCT00104650) included men and women with solid tumors and bone metastases or multiple myeloma with evidence of high levels of bone resorption (uNTX levels > 50 nM BCE/mM creatinine) despite previous treatment with IV BPs for 8 weeks or more.17 In both studies, patients were excluded if they had more than two prior SREs, osteonecrosis or osteomyelitis of the jaw (current or past), planned oral surgery, radiotherapy to bone less than 2 weeks before randomization, or evidence of impending fracture in weight-bearing bones.

The studies were approved by the institutional review board or ethics committee for each site. All patients provided written informed consent.

Study designs and treatments

Both studies were randomized, active-controlled, multicenter phase 2 trials comparing different doses of subcutaneous denosumab with IV BPs 16, 17 (Fig. 1). Patients were randomly assigned to receive IV BP therapy every 4 weeks or the assigned dose of denosumab [denosumab every 4 weeks (30, 120, or 180 mg) or denosumab every 12 weeks (60 or 180 mg)]. Randomization in both studies was stratified. In the study of bisphosphonate-naive patients, patients were stratified by type of antineoplastic therapy (i.e., hormonal therapy or chemotherapy); in the study of previously treated patients, stratification was by cancer type (i.e., prostate cancer, breast cancer, multiple myeloma/other solid tumors) and baseline uNTX (50 to 100 or >100 nmol/L/mM creatinine). IV BPs administered were commercially available agents such as zoledronic acid, ibandronate, or pamidronate, selected at the investigator's discretion and administered according to country-specific labeling. Patients were treated for 25 weeks and were followed for up to 32 weeks after the treatment phase was completed. In study 20040114, patients had the option of entering a 2-year ongoing extension study. All patients were instructed to take daily supplements of calcium (500 mg) and vitamin D (≥400 IU).

Figure 1. Study designs.

Initial and periodic study assessments included a medical history and physical examination, recording of vital signs, electrocardiograms, radiographic evaluations of the spine, and laboratory assessments, including hematology, serum chemistry, and measurement of bone turnover markers. Concurrent antineoplastic or hormonal therapy was permitted during study treatment at the investigators' discretion as long as no changes in regimens or agents were planned within 4 weeks before or after randomization.

Endpoints

Key efficacy endpoints analyzed for this report include the median percentage change from baseline in uNTX corrected for creatinine (uNTX/Cr) at 25 weeks, the median percentage change from baseline in other bone turnover markers (sCTX, BSAP, TRAP-5b, P1NP, and osteocalcin), the proportion of patients experiencing SREs after 25 weeks, and the time to the first SRE. SREs were defined as pathologic fracture, either vertebral or nonvertebral; spinal cord compression; surgery to bone; or radiation to bone, including the use of radioisotopes.

Endpoints reported previously include the time to a reduction of uNTX > 65% at weeks 13 and 25 in the IV BP-naive patients16, 18 and the time to uNTX < 50 nM BCE/mM creatinine at weeks 13 and 25 in the patients previously treated with IV BPs.17 The cutoff value for uNTX < 50 nM BCE/mM creatinine was selected based on previously published data suggesting that uNTX > 50 nM BCE/mM creatinine increased the risk for SREs, cancer progression, and death.14 Safety endpoints included the incidence of adverse events (AEs), changes in laboratory values (including calcium levels and renal function), and the formation of antibodies.

Statistical analysis

Summary statistics were calculated for continuous and categorical variables. Kaplan-Meier analyses were conducted for the times to the occurrence of the first SRE.

Analyses of efficacy endpoints included only data from the 25-week treatment phase. Safety analyses included data from both the 25-week treatment phase and the 32-week off-treatment follow-up phase for a total of 57 weeks. For patients from study 20040114 who enrolled in the extension study, data were censored at the time they entered the extension.

Results

Demographics and baseline characteristics

All the patients in the IV BP-naive study were women with breast cancer and bone metastases. In the study of patients previously treated with IV BPs, the population was evenly distributed between men and women, with breast, prostate, and other solid tumors and multiple myeloma similarly distributed between treatment groups (Table 1). Prostate cancer was the most common tumor type (45%) in this second study.

Table 1. Baseline Demographics and Disease Characteristics of Study Populations

	IV BP-naive patients		Patients previously treated with IV BPs
	IV BP (n = 43)	All denosumab (n = 212)	IV BP (n = 37)	All denosumab (n = 74)
IV BP, intravenous bisphosphonate; SD, standard deviation; ECOG, Eastern Cooperative Oncology Group; GFR, glomerular filtration rate; SREs, skeletal-related events; uNTX, urinary N-telopeptide; Cr, creatinine; Q1, first quartile; Q3, third quartile; sCTX, serum C-telopeptide.
Sex, n (%)
Women	43 (100)	212 (100)	18 (49)	38 (51)
Men	0 (0)	0 (0)	19 (51)	36 (49)
Age, mean, years (SD)	52 (11)	58 (11)	62 (12)	63 (12)
ECOG status, n (%)
0	19 (44)	135 (64)	9 (24)	15 (20)
1	20 (47)	70 (33)	23 (62)	36 (49)
2	4 (9)	6 (3)	5 (14)	21 (28)
3	0 (0)	1 (1)	0 (0)	0 (0)
Unknown	0 (0)	0 (0)	0 (0)	2 (3)
Tumor type, n (%)
Breast cancer	43 (100)	212 (100)	16 (43)	30 (40)
Prostate cancer	0 (0)	0 (0)	17 (46)	33 (45)
Multiple myeloma	0 (0)	0 (0)	3 (8)	6 (8)
Other solid tumor	0 (0)	0 (0)	1 (3)	5 (7)
Time since original diagnosis, years, median (min, max)	3.1 (0.0, 15.6)	3.3 (0.0, 32.8)	3.1 (0.2, 13.2)	3.7 (0.1, 21.0)
Time since bone metastases, months, median (min, max)	2.1 (0.3, 26.9)	1.9 (0.1, 206.2)	10.8 (0.0, 103.2)	7.2 (0.0, 237.6)
Bone metastases > 2, n (%)	34 (79)	156 (74)	34 (92)	68 (92)
Previous SREs ≥ 1	15 (35)	72 (34)	16 (43)	47 (64)
Estimated GFR (mL/min/1.73 m^2), median (min, max)	98.7 (51.1, 160.2)	88.8 (29.0, 200.9)	101.1 (48.5, 182.8)	86.1 (30.8, 275.0)
Baseline uNTX/Cr (nM/mM), median (Q1, Q3)	49.1 (22.7, 111.7)	46.1 (25.3, 103.3)	103.2 (58.0, 145.7)	107.1 (48.6, 211.2)
Baseline sCTX (ng/mL), median (Q1, Q3)	0.6 (0.4, 0.7)	0. 6 (0.3, 0.8)	0.8 (0.4, 1.3)	0.8 (0.5, 1.5)
Breast cancer patients receiving hormone therapy, n (%)	23 (54)	122 (58)	13 (81)	23 (77)
Prostate cancer patients receiving hormone therapy, n (%)	—	—	14 (82)	25 (76)

In the IV BP-naive study, patients generally were older in the denosumab groups than in the IV BP group (mean age 58 versus 52 years). In the study of patients previously treated with IV BPs, the study population tended to be older than those in the IV BP-naive study, with a mean age of 63 years (see Table 1). In both studies, bone turnover marker levels were similar in the IV BP and denosumab groups (see Table 1). Zoledronic acid was the most commonly used BP (by 91% of patients in the BP-naive study and 81% of patients in the previous-BP study). Approximately half the patients in the study of BP-naive patients and more than three-quarters of breast and prostate cancer patients in the study of patients previously treated with IV BPs were receiving hormone therapy at baseline (see Table 1).

Bone turnover

Patients treated with denosumab experienced a rapid reduction in bone turnover marker levels in both studies. In the study of IV BP-naive patients, the median percent change in uNTX/Cr at 25 weeks was –75% for the denosumab groups and –71% for the IV BP group (Fig. 2A). In the study of patients previously treated with IV BPs, denosumab-treated patients experienced a median reduction in uNTX/Cr of 80% compared with a reduction of 56% for those in the IV BP group (see Fig. 2B). Results were similar for the subsets of patients with breast and prostate cancer (see Fig. 2C, D). Reductions in uNTX were greater in the IV BP treatment group in the BP-naive study than in the study of patients who had high levels of uNTX despite previous BP therapy (see Fig. 2). Patients previously treated with IV BPs reached a uNTX level of less than 50 nM BCE/mM creatinine in a median 9 days for the denosumab groups and 65 days for the IV BP group; among IV BP-naive patients, the median time to reach uNTX less than 50 nM BCE/mM creatinine was 9 days for the denosumab groups and 8 days for the IV BP group. The receipt of hormone therapy or chemotherapy at baseline had no significant effect on changes in uNTX levels (data not shown). Denosumab-induced suppression of bone turnover also was demonstrated in the median reductions of serum levels of other markers analyzed (Fig. 3). Among patients previously treated with IV BPs, the difference in the median change of the osteoclast marker TRAP-5b at week 25 was substantial between the denosumab and IV BP treatment arms (–73% for denosumab versus –11% for IV BP).

Figure 2. Median reduction in uNTX/Cr from baseline through 25 weeks.

Figure 3. Median percentage change in secondary bone turnover markers at 25 weeks.

Skeletal-related events (SREs)

The incidence of SREs was low in both treatment groups across both studies. Among IV BP-naive patients, 12% of patients in the denosumab groups and 16% of patients in the IV BP group experienced a first on-study SRE over 25 weeks. Among patients previously treated with IV BPs, the rate of first on-study SRE was lower in the denosumab groups (8%) than the IV BP group (17%), and IV BP-treated patients experienced SREs earlier than denosumab-treated patients.17 Among IV BP-naive patients, the Kaplan-Meier curves of the time to first on-study SRE were similar in both treatment groups.17, 18

Safety

No unexpected changes in calcium, creatinine, liver enzymes, or electrolytes were reported in either study, and the overall profile of changes in serum calcium was similar in both treatment groups in both studies. The median changes in serum calcium did not exceed 0.05 mmol/L in either group. Denosumab treatment had no apparent effect on renal function in either study.

The rates of AEs were similar between treatment groups in both studies (Table 2), and the events reported were consistent with a population of patients undergoing treatment for advanced cancer. Rates of treatment-related events of Common Terminology Criteria for Adverse Events (CTCAE) grades 3, 4, or 5 were low and similar in both studies (see Table 2). No treatment-related deaths were reported. Most deaths were from disease progression, and rates of death were similar between treatment groups in both studies. No cases of osteonecrosis of the jaw and no neutralizing antibodies to denosumab were reported during the 25-week treatment and 32-week follow-up periods.

Table 2. Summary of Adverse Events Through Week 57

	IV BP-naive patients		Patients Previously Treated with IV BPs
	IV BP (n = 43)	All denosumab (n = 211)	IV BP (n = 35)	All denosumab (n = 73)
a One serious treatment-related AE (pyrexia) was recorded in the 120-mg every 4 weeks denosumab group, but it was determined to be unrelated to treatment after the study database was locked. IV BP, intravenous bisphosphonate; Q4W, every 4 weeks; Q12W, every 12 weeks; CTCAE, Common Terminology Criteria for Adverse Events.
Number of patients reporting any AEs, n (%)	41 (95)	200 (95)	34 (97)	70 (96)
AEs of CTCAE grades 3, 4, or 5, n (%)	20 (47)	93 (44)	25 (71)	40 (55)
Number of patients reporting serious AEs	15 (35)	75 (36)	19 (54)	37 (51)
Treatment-related AEs, n (%)	13 (30)	45 (21)	3 (9)	19 (26)
Treatment-related serious AEs, n (%)	0 (0)	0 (0)a	0 (0)	1 (1)
Withdrawals from study because of AEs, n (%)	1 (2)	5 (2)	3 (9)	4 (6)
Deaths, n (%)	8 (19)	32 (15)	12 (34)	23 (32)
Treatment-related deaths	0 (0)	0 (0)	0 (0)	0 (0)
Adverse events reported by 10% or more of patients receiving denosumab in either study
Nausea	10 (23)	47 (22)	7 (20)	17 (23)
Vomiting	8 (19)	36 (17)	6 (17)	7 (10)
Diarrhea	7 (16)	35 (17)	4 (11)	10 (14)
Asthenia	12 (28)	34 (16)	7 (20)	15 (20)
Back pain	4 (9)	30 (14)	5 (14)	8 (11)
Headache	8 (19)	28 (13)	1 (3)	5 (7)
Fatigue	5 (12)	28 (13)	4 (11)	8 (11)
Bone pain	8 (19)	26 (12)	12 (34)	21 (29)
Constipation	7 (16)	26 (12)	6 (17)	16 (22)
Anemia	2 (5)	23 (11)	8 (23)	17 (23)
Arthralgia	13 (30)	24 (11)	1 (3)	6 (8)
Pain in extremity	8 (19)	21 (10)	3 (9)	7 (10)
Pyrexia	9 (21)	18 (9)	1 (3)	7 (10)
Cough	7 (16)	18 (9)	4 (11)	5 (7)
Peripheral edema	6 (14)	14 (7)	1 (3)	11 (15)
Dyspnea	5 (12)	12 (6)	4 (11)	7 (10)
Paresthesia	4 (9)	11 (5)	3 (9)	10 (14)
Thrombocytopenia	0 (0)	5 (2)	2 (6)	9 (12)

Discussion

Elevated bone turnover markers are associated with disease progression and poor prognosis in breast cancer, prostate cancer, and other solid tumors with bone metastases. Inhibition of osteoclast function, as measured by decreases in bone resorption markers, results in fewer skeletal complications and a more favorable prognosis.1 In these two studies, denosumab reduced levels of uNTX/Cr and other bone turnover markers such as sCTX and TRAP-5b consistently regardless of tumor types, prior history of SRE, and prior IV BP exposure. Notably, among patients whose uNTX levels remained high despite previous IV BP therapy, denosumab rapidly decreased uNTX and other markers of bone turnover, whereas continuation of BP treatment led only to a progressive and lower decrease in uNTX levels. This slower effect may be due in part to the interference of changing antineoplastic treatments over time. The difference in TRAP-5b levels between denosumab-treated patients and those continuing IV BPs is especially striking and suggests the persistence of functioning osteoclasts despite BP treatment but whose activity can be suppressed when treatment is switched to denosumab. These results demonstrate the biologic activity of denosumab. They also confirm a mechanism of action for denosumab that is distinct from that of BPs. BPs are intercalated into bone and inhibit the osteoclast's ability to resorb bone. Denosumab, by inhibiting the interaction of RANK with its ligand, prevents the formation, maturation, and survival of osteoclasts. In studies of animal models treated with the RANKL inhibitor osteoprotegerin (OPG), surviving osteoclasts were observed after BP treatment but not after treatment with OPG.19, 20 In bone biopsies from patients who died with multiple bone metastases, surviving osteoclasts were observed after BP treatment but not after treatment with denosumab.21 Because it effectively inhibits osteoclast formation, maturation, and survival, denosumab may be effective for patients in whom osteoclasts persist or are still formed despite treatment with BPs. Persistent osteoclast activity was confirmed in the patients previously treated with BP by the fact that TRAP-5b levels did not decrease after further BP treatment. The clinical implications of these findings may be important but remain to be demonstrated.

Among IV BP-naive patients, the incidence of SREs was similarly low after IV BP and denosumab treatment. Among patients with prior exposure to IV BPs, the incidence of SREs was lower in the denosumab group than in patients on continuing IV BPs. The ability of denosumab to further suppress bone turnover markers, particularly uNTX and the osteoclast biomarker TRAP-5b, in patients already receiving IV BP but continuing to exhibit evidence of a high bone resorption rate may account for the lower rate of SREs in the second study. Because of the relatively small sizes of the populations, larger phase 3 studies are ongoing to provide a more accurate estimate of the effect of denosumab treatment on the risk of SREs.

The rates or types of AEs, deaths, and serious adverse events appeared to be similar between groups in both studies, and no additional adverse events surfaced related to switching from BP to denosumab.

Growing evidence indicates that a bone-targeting strategy is likely to be a valuable and promising approach in patients with bone metastases.22, 23 Results of the current studies add to this evidence, demonstrating that denosumab may provide an important therapeutic advance in the prevention of skeletal complications of bone metastases. Phase 3 studies of denosumab in patients with solid tumors and multiple myeloma are ongoing.

Disclosures

JJB has received consultancy and lecture fees from Amgen, Inc., and Novartis. AL has received consultancy fees and honoraria from Amgen, Inc. AL also has received honoraria, consulting fees, research funds, and payments for expert testimony from Novartis, Inc. JG has received research funding from Amgen, Inc., Novartis, Inc., and Roche. GGS has received honoraria from Amgen, Inc. GG and HY are employees of Amgen, Inc., and have received stocks/stock options from Amgen, Inc. KF has received consultancy fees and/or honoraria from Amgen, Inc., AstraZeneca, Sanofi-Aventis, Novartis, Ipsen-Beaufour, Pharmion, Bristol Myers Squibb, and Takeda. KF also has received research funding from Amgen, Inc.

Acknowledgements

We acknowledge the medical writing assistance of Ting Chang and Sue Hudson on behalf of Amgen, Inc. This study was supported by Amgen, Inc., Thousand Oaks, CA, USA.