Abstract
Context:
Hypercalcemia of malignancy (HCM) in patients with advanced cancer is often caused by excessive osteoclast-mediated bone resorption. Patients may not respond to or may relapse after iv bisphosphonate therapy.
Objective:
We investigated whether denosumab, a potent inhibitor of osteoclast-mediated bone resorption, reduces serum calcium in patients with bisphosphonate-refractory HCM.
Design, Setting, and Participants:
In this single-arm international study, participants had serum calcium levels corrected for albumin (CSC) >12.5 mg/dL (3.1 mmol/L) despite bisphosphonates given >7 and ≤30 days before screening.
Intervention:
Patients received 120 mg sc denosumab on days 1, 8, 15, and 29 and then every 4 weeks.
Main Outcome Measures:
The primary endpoint was the proportion of patients with CSC ≤11.5 mg/dL (2.9 mmol/L) (response) by day 10. Secondary endpoints included response by visit, duration of response, and the proportion of patients with a complete response (CSC ≤10.8 mg/dL [2.7 mmol/L]) by day 10 and during the study.
Results:
Patients (N = 33) had solid tumors or hematologic malignancies. By day 10, 21 patients (64%) reached CSC ≤11.5 mg/dL, and 12 patients (33%) reached CSC ≤10.8 mg/dL. During the study, 23 patients (70%) reached CSC ≤11.5 mg/dL, and 21 patients (64%) reached CSC ≤10.8 mg/dL. Estimated median response duration was 104 days. The most common serious adverse events were hypercalcemia worsening (5 patients, 15%) and dyspnea (3 patients, 9%).
Conclusions:
In patients with HCM despite recent iv bisphosphonate treatment, denosumab lowered serum calcium in 64% of patients within 10 days, inducing durable responses. Denosumab may offer a new treatment option for HCM.
Hypercalcemia of malignancy (HCM) is a complication of patients with advanced cancer. It is characterized by elevated serum calcium and indicates a poor prognosis (1, 2). The estimated prevalence of HCM in cancer patients in the United States in 2012 was 2.7% and varies with tumor type, ranging from 1.5% for prostate cancer to 9.5% for multiple myeloma (3). Symptoms include nausea, vomiting, abdominal pain, bone pain, fatigue, and confusion. Ultimately, HCM can result in renal failure, coma, and death, with an estimated 50% survival of 30 days regardless of treatment (4).
HCM is often caused by tumor-induced bone resorption mediated by increased osteoclast activity, through either a humoral mechanism or local cancer-induced osteolysis. Humoral HCM is characterized by the systemic secretion of PTHrP by malignant cells, which promotes increased bone resorption by osteoclasts (2, 5–8). Ectopic secretion of authentic PTH by the tumor itself is a rare phenomenon (2). Local osteolytic bone resorption occurs in areas near malignant cell invasion, where tumor cells secrete osteoclast-activating cytokines (macrophage inflammatory protein-1α, IL-1 and -6, PTHrP, or receptor activator of nuclear factor-κB ligand [RANKL]) that may enter the local bone component and/or systemic circulation. To date, no guidelines are available from professional societies regarding the treatment of HCM. Definitive treatment of HCM relies upon effective treatment of the underlying malignancy. However, this is not always possible, and treatment of HCM is required to palliate patient symptoms. Initial therapy for HCM includes saline to correct the volume depletion, which will increase the glomerular filtration rate and renal excretion of calcium. Once euvolemia is established, a loop diuretic can be considered to encourage further calciuresis, although clinical results vary (9). Calcitonin may be used for its fast onset of action and its lack of nephrotoxic effects, but its value is limited by short duration of efficacy, small reduction of calcium level, and potential development of tachyphylaxis (1).
For hypercalcemia that persists despite initial interventions, iv bisphosphonates are the treatment of choice (2). In a pooled analysis of 2 pivotal studies comparing zoledronic acid to pamidronate in patients with HCM and corrected serum calcium (CSC) levels of ≥12.0 mg/dL (3.0 mmol/L), a complete response to treatment (defined as CSC ≤10.8 g/dL [2.7 mmol/L]) was reported in 88% and 70% of patients who received zoledronic acid 4 mg or pamidronate 90 mg, respectively (10, 11). In these 2 studies, 24% of patients treated with a single dose of zoledronic acid 4 mg or pamidronate relapsed (defined as CSC ≥11.6 mg/dL [2.9 mmol/L]) within 56 days. Another 22% responded incompletely to treatment (defined as failing to achieve CSC reduction from baseline of at least 0.2 mg/dL [0.05 mmol/L] by day 4 or 1.0 mg/dL [0.25 mmol/L] by day 8, or CSC ≥11.6 mg/dL [2.9 mmol/L] by day 19) (10). Bisphosphonate therapy requires renal monitoring and may not be appropriate for patients with severe renal impairment (12, 13). New treatment alternatives are needed to help patients for whom current therapies are unsatisfactory.
Denosumab, a fully human monoclonal antibody, binds RANKL to inhibit the formation, function, and survival of osteoclasts, the cells responsible for bone resorption. A preclinical study used osteoprotegerin, an endogenous decoy receptor that binds and neutralizes RANKL, to evaluate the role of osteoclast-mediated hypercalcemia in 2 murine models of humoral HCM. In one model, mice were inoculated with colon adenocarcinoma cells, and in the other model, mice were injected with high-dose PTHrP twice daily. In both models, RANKL inhibition with a single injection of osteoprotegerin caused a more rapid reversal of established hypercalcemia and longer sustained hypercalcemia suppression than high-dose bisphosphonates (14). In another study, daily administration of a RANKL inhibitor with RANK-Fc to animals bearing an sc lung squamous cell carcinoma also profoundly inhibited osteoclastic bone resorption and prevented hypercalcemia (15). Currently, denosumab is approved by the U.S. Food and Drug Administration at 120 mg every 4 weeks for patients with bone metastases due to solid tumors and patients with giant cell tumor of bone not amenable to curative surgery.
In phase 3 studies, denosumab therapy was associated with a lower incidence of skeletal-related events or HCM in patients with advanced malignancies with bone metastases compared with zoledronic acid (16–18). In an integrated analysis of patients with multiple myeloma or solid tumors (excluding prostate cancer) and bone metastases, HCM occurred in 32 patients (1.7%) treated with denosumab and 52 patients (2.7%) treated with zoledronic acid (19). Denosumab also prolonged the time to first hypercalcemia by 37% compared with zoledronic acid (hazard ratio 0.63, 95% confidence interval [CI], 0.41–0.98, P = .04).
We evaluated denosumab for treatment of HCM in patients who remained significantly hypercalcemic despite recent iv bisphosphonate treatment. Bisphosphonate-refractory is defined as when the CSC does not decrease below or equal to 11.5 mg/dL 7 to 30 days after treatment with iv bisphosphonate. Results from a prespecified interim analysis of 15 patients have been previously reported (20). This report presents results of the primary analysis of all 33 patients from this study, where the primary analysis is defined as all applicable efficacy and safety parameters from the first 57 days of treatment and assessments.
Subjects and Methods
Study design and participants
In this open-label, single-arm, multicenter, international, phase 2 study, the first patient was enrolled on November 16, 2009; the cutoff date for this primary analysis was September 13, 2012, prespecified as the date by which all patients had the opportunity to receive at least 1 dose of denosumab and complete the specified 57-day assessments. Participants were adult patients with cancer and hypercalcemia refractory to iv bisphosphonates, defined as CSC levels >12.5 mg/dL (3.1 mmol/L) despite iv bisphosphonate treatment within 7 to 30 days before screening CSC evaluation. This CSC level corresponds to a grade 3 (severe) adverse event on the Common Terminology Criteria for Adverse Events (CTCAE) scale (21). At least 7 days were required since last bisphosphonate dose because maximal effects may take 7 days to manifest, consistent with the prescribing information for zoledronic acid (13). Patients had adequate organ function, defined as serum aspartate aminotransferase ≤5 × the upper limit of normal (ULN), serum alanine aminotransferase ≤5 × ULN, and serum total bilirubin ≤ 2 × ULN. Patients were excluded if they had benign hyperparathyroidism, hyperthyroidism, or adrenal insufficiency or were on dialysis. Patients were also excluded if they were receiving treatment before screening with thiazides, calcitonin, mithramycin, or gallium nitrate within the window of expected therapeutic effect for each drug, as determined by the treating physician, or cinacalcet within 4 weeks before screening. Concurrent iv fluids, steroids, and chemotherapy were permitted. The independent ethics committee or institutional review board at each site approved the protocol, and each patient provided written informed consent before participation.
Intervention
Patients received sc denosumab 120 mg on days 1, 8, 15, and 29 and every 4 weeks thereafter. This regimen is based on the dose used for patients with bone metastases (120 mg every 4 weeks), with 2 additional administrations given on days 8 and 15 to reach steady-state denosumab concentrations and reduce serum calcium levels quickly. The rapid achievement of effective serum concentrations of denosumab using this dose and schedule was demonstrated in a clinical study of patients with giant cell tumor of bone (22). This regimen, recently approved in the United States and Canada for the treatment of giant cell tumor of bone (13, 23), was well tolerated in clinical studies of more than 300 patients (22, 24).
Denosumab was discontinued if CSC was >12.5 mg/dL after 4 denosumab doses or by study day 57 (unless investigator considered treatment to be of continued clinical benefit) or if the patient withdrew consent. On-study blood samples for analysis of CSC and other blood chemistry were collected on days 1, 2, 4, 8, 10, 15, 19, 23, and 29 and then weekly until day 57 and monthly thereafter until the end of the study. Screening blood samples were analyzed by local laboratories to determine eligibility; on-study blood samples were analyzed by a central laboratory. Urine samples were collected at screening and on days 8, 29, and 57 and monthly thereafter for analysis of urinary-N-telopeptide corrected for urinary creatinine (uNTx/uCr) and urinary calcium corrected for urinary creatinine.
Study outcomes
Primary endpoint
The primary endpoint was the proportion of patients with a response, prespecified as CSC ≤11.5 mg/dL (2.9 mmol/L; CTCAE grade 0 or 1, corresponding to normal or mildly elevated calcium) within 10 days after the first dose of denosumab. CSC was calculated as: total serum calcium in milligrams per deciliter + [0.8 × (4 − serum albumin in grams per deciliter)].
Secondary endpoints
Key prespecified secondary efficacy endpoints included response by visit, duration of response (defined as the number of days from the first occurrence of CSC ≤11.5 mg/dL to the last continuous CSC value ≤11.5 mg/dL), and the proportion of patients who experienced a complete response (CSC ≤10.8 mg/dL [2.7 mmol/L]) by visit. Other secondary endpoints included time to response, time to complete response, duration of complete response, time to relapse/nonresponse, and changes in CSC from baseline. Endpoint definitions were consistent with previous studies of iv bisphosphonates in HCM (10, 11).
Other endpoints
Key exploratory efficacy endpoints included changes in CSC in relation to PTHrP and uNTx/uCr. Results were also analyzed for subsets of patients based on demographic factors, the presence of bone metastases, and baseline values of PTHrP and creatinine clearance (CrCl). Relief of symptoms by day 10 was also evaluated in a prespecified analysis. Safety endpoints included the type, frequency, and severity of adverse events and changes in laboratory values.
Statistical methods
Sample size was prespecified to be approximately 33 patients, with 2-sided 95% exact CIs to be estimated for patient incidence of various response rates. Efficacy and safety analyses were performed on all patients who received at least 1 dose of denosumab. The proportions of subjects who achieved response and complete response were provided with the associated exact 95% CIs. Categorical data were reported as numbers or percentages. Continuous data were reported with descriptive statistics (mean, SD; medians with 25th and 75th percentiles [Q1, Q3], or minimum and maximum). Time-to-event variables were reported using Kaplan-Meier point estimates with 95% CI (using bootstrap methods) and median (Q1, Q3) when possible.
Results
Patient disposition
At the primary analysis cutoff date, 35 patients had been screened; 33 patients were enrolled, received at least 1 dose of denosumab, and were included in the efficacy and safety analyses. Two patients were continuing to receive denosumab treatment, and 31 (94%) had discontinued the study (Figure 1). The most common reasons for discontinuation were death (10 patients, 30%), adverse events (8 patients, 24%), and progression of hypercalcemia (5 patients, 15%). The adverse events that led to discontinuation were progression of breast cancer in 2 patients; progression of bladder cancer, head and neck cancer, non-small-cell lung cancer, or renal cell carcinoma (n = 1 for each); and hepatic failure and hypoxia (n = 1 for each). None of these events was considered denosumab-related by the investigator.
Figure 1.
Consolidated Standards of Reporting Trials (CONSORT) flowchart: patient disposition. At the primary analysis cutoff date, 35 patients had been screened; 33 patients were enrolled, received at least 1 dose of denosumab, and were included in the efficacy and safety analyses. Two patients were continuing to receive denosumab treatment, and 31 (94%) had discontinued the study.
Baseline characteristics
Baseline characteristics are reported in Table 1. Twenty-four patients (73%) had advanced solid tumors, and 9 (27%) had advanced hematologic malignancies. Most patients (76%) had poor performance status (Eastern Cooperative Oncology Group status ≥2), and 30 patients (91%) had metastatic disease, including 13 (39%) with bone metastases. The median CSC level at baseline measured by the central laboratory was 13.7 (range, 11.9–17.3) mg/dL. Nineteen patients reported 48 symptoms at baseline, including 9 patients with 12 symptoms of cognitive impairment that included lethargy, confusion, and depressed consciousness (Table 1). The most common (>10%) symptoms at baseline were fatigue, anorexia, nausea, lethargy, and constipation. Patients in the study had received a median of 5 doses (Q1, Q3: 3, 11) and a maximum of 42 doses of iv bisphosphonates over a median of 4 months (Q1, Q3: 2, 11) before enrollment; 28 patients (85%) had received at least 2 doses of iv bisphosphonates, and 10 patients (30%) had received more than 10 doses. The median time from last iv bisphosphonate treatment to study enrollment was 17 days (Q1, Q3: 13, 22). Patients received a median of 4 doses (Q1, Q3: 2, 5) of denosumab, and a maximum of 25 doses. The median time on study was 56 days (Q1, Q3: 18, 79). Nine patients did not receive at least 1 scheduled dose; reasons included low or normal calcium (2 patients), adverse events (5 patients), hip replacement surgery (1 patient), and investigator's discretion (1 patient).
Table 1.
Baseline Characteristics
| Characteristic | Denosumab Treatment Group, N = 33 |
|---|---|
| Sex, male, n (%) | 21 (64) |
| Ethnic group/race, n (%) | |
| White or Caucasian | 23 (70) |
| Black or African American | 7 (21) |
| Other | 3 (9) |
| Age, median (range), y | 63 (22–89) |
| ECOG performance status, n (%) | |
| 0–1 | 8 (24) |
| 2–4 | 25 (76) |
| Primary tumor type (%) | |
| Solid tumors | 24 (73) |
| Breast | 6 (18) |
| Neuroendocrine | 4 (12) |
| Non-small-cell lung cancer | 3 (9) |
| Renal cell | 3 (9) |
| Head and neck | 2 (6) |
| Adenocarcinoma with unknown primary site | 1 (3) |
| Bladder | 1 (3) |
| Liver | 1 (3) |
| Ovarian | 1 (3) |
| Small-cell lung cancer | 1 (3) |
| Soft tissue sarcoma | 1 (3) |
| Hematologic malignancies | 9 (27) |
| Myeloma | 5 (15) |
| Chronic lymphocytic leukemia with Richter's transformation | 2 (6) |
| Non-Hodgkin lymphoma | 2 (6) |
| Metastatic disease at baseline, n (%) | 30 (91) |
| Presence of bone metastasis at baseline, n (%) | 13 (39) |
| Months from initial cancer diagnosis to enrollment, median (Q1, Q3) | 30 (10, 84) |
| Months of prior bisphosphonate use, median (range) | 4 (1–41) |
| Number of doses of intravenous bisphosphonates, median (range) | 5 (1–42) |
| Days from last intravenous bisphosphonate treatment to enrollment, median (Q1, Q3) | 17 (13, 22) |
| Patients reporting symptoms attributed to hypercalcemia at baseline, n (%)a | 19 (58) |
| Fatigue | 9 (27) |
| Anorexia | 5 (15) |
| Nausea | 4 (12) |
| Constipation | 4 (12) |
| Lethargy | 4 (12) |
| Otherb | 22 (56) |
| Calcium level (albumin corrected), median (range), mg/dLc,d | 13.7 (11.9–17.3) |
| PTHrP level, median (range), pmol/Le | 4.2 (0.5–24.0) |
| Calculated creatinine clearance, median (range), mL/minf | 76 (13–311) |
| uNTx/uCr level, median (range), nmol/mmolg | 76.9 (7–425) |
Abbreviation: ECOG, Eastern Cooperative Oncology Group.
Some patients experienced more than 1 symptom.
Other symptoms included confusion (reported in 3 patients); depressed level of consciousness, polyuria, and vomiting (2 patients each); and abdominal pain, decreased mental acuity, dry mouth, dyspnea, general weakness, insomnolence, dizziness, listlessness/muscle weakness, polyuropolydipsic syndrome, psychomotor retardation, retching, urinary frequency, and weight loss (1 patient each).
If albumin was <4.0 g/dL, CSC was calculated as total serum calcium in mg/dL + [0.8 × (4 − serum albumin in g/dL)]. If albumin was ≥4.0 g/dL, corrected serum calcium was equal to total calcium in mg/dL.
The minimum baseline value is below the specified minimum for inclusion because local laboratory values were used to screen patients for study enrollment, and locally measured values may have differed from central laboratory values. For 6 patients, baseline central laboratory data were not available, and their screening values were used.
A PTHrP value ≥2 pmol/L is considered elevated. Baseline PTHrP data were missing for 9 patients.
Cockcroft-Gault formula. Baseline CrCl data were missing for 5 patients.
Baseline uNTx/uCr data were missing for 4 patients.
Efficacy results
Reduction of serum calcium
By day 10, 21 patients (64%) met the primary endpoint, responding to denosumab treatment with CSC levels ≤11.5 mg/dL (Table 2). Also by day 10, 12 patients (36%) had a complete response (CSC ≤10.8 mg/dL). Over the course of the study, 23 patients (70%) had a response and 21 patients (64%) had a complete response. The estimated median time to response was 9 days (95% CI, 5–19), and the estimated median duration of response was 104 days (95% CI, 9 to not estimable). Of the 23 patients with a response, the estimated median time to relapse was 114 days (95% CI, 19.0 to not estimable). The median CSC level remained below 11.5 mg/dL through day 57 (Figure 2). After initially responding to denosumab, 10 patients relapsed, with CSC levels rising above 11.5 mg/dL; of these patients, 8 (80%) responded again while on denosumab treatment. The estimated median time to complete response was 23 days (95% CI, 11–43), and the estimated median duration of complete response was 34 days (95% CI, 1–134).
Table 2.
Denosumab Efficacy in HCMa
| Patients Experiencing a Treatment Response (CSC ≤11.5 mg/dL) |
Patients Experiencing a Complete Response (CSC ≤10.8 mg/dL) |
|||
|---|---|---|---|---|
| By Day 10 | Over the Course of the Study | By Day 10 | Over the Course of the Study | |
| Overall population, N = 33 | 21 (64) | 23 (70) | 12 (36) | 21 (64) |
| Patient subgroups (characteristics at baseline) | ||||
| Gender | ||||
| Women, n = 12 | 5 (42) | 7 (58) | 4 (33) | 6 (50) |
| Men, n = 21 | 16 (76) | 16 (76) | 8 (38) | 15 (71) |
| Age | ||||
| <65 y, n = 19 | 12 (63) | 12 (63) | 6 (32) | 11 (58) |
| ≥65 y, n = 14 | 9 (64) | 11 (79) | 6 (43) | 10 (71) |
| Racial group | ||||
| Caucasian, n = 23 | 14 (61) | 15 (65) | 9 (39) | 14 (61) |
| Non-Caucasian, n = 10 | 7 (70) | 8 (80) | 3 (30) | 7 (70) |
| Bone metastases | ||||
| Bone metastases, n = 13 | 7 (54) | 8 (62) | 3 (23) | 6 (46) |
| No bone metastases, n = 20 | 14 (70) | 15 (75) | 9 (45) | 15 (75) |
| Symptoms | ||||
| Symptomatic hypercalcemia, n = 19 | 15 (79) | 16 (84) | 9 (47) | 14 (74) |
| Asymptomatic hypercalcemia, n = 14 | 6 (43) | 7 (50) | 3 (21) | 7 (50) |
| CSC level at baseline | ||||
| >13.5 mg/dL, n = 19 | 12 (63) | 13 (68) | 7 (37) | 13 (68) |
| ≤13.5 mg/dL, n = 14 | 9 (64) | 10 (71) | 5 (36) | 8 (57) |
| PTHrP level at baselineb | ||||
| >2 pmol/L, n = 19 | 13 (68) | 14 (74) | 11 (58) | 14 (74) |
| ≤2 pmol/L, n = 5 | 3 (60) | 4 (80) | 1 (20) | 4 (80) |
| Renal function | ||||
| CrCl <60 mL/min, n = 11 | 6 (55) | 8 (73) | 3 (27) | 8 (73) |
| CrCl ≥60 mL/min, n = 17 | 10 (59) | 10 (59) | 7 (41) | 10 (59) |
Results are shown as n (%).
A PTHrP level >2 pmol/L is associated with PTHrP-mediated HCM. Baseline PTHrP data were missing for 9 patients. The small size of population subgroups warrants caution in the interpretation of data and precludes further statistical analysis.
Figure 2.
CSC at each study visit (median [Q1, Q3]). N = the number of patients who received at least 1 dose of denosumab, and n = the number of patients for whom a CSC value was available at each time point.
In subgroup analyses, denosumab reduced CSC for patients in the study regardless of gender, age, racial group, the presence or absence of bone metastases, symptomatic vs asymptomatic hypercalcemia, baseline levels of PTHrP, or renal impairment (Table 2). In nearly all subgroups, at least half of patients experienced a response by day 10. Little correlation was observed between PTHrP and CSC, and changes in CSC in relationship to baseline PTHrP levels showed no consistent pattern. Patients with impaired renal function (CrCl <60 mL/min) responded to treatment in similar proportions as patients with normal renal function (CrCl ≥60 mL/min) (55% vs 59%). Responses were similar based on the number of prior doses of iv bisphosphonates received or the degree of baseline bone turnover as indicated by uNTx/uCr level. Of the 17 patients with a baseline uNTx/uCr ≥50 nmol/mmol despite recent bisphosphonate treatment, 14 (82%) had a response by day 10. Reductions in serum calcium and reductions in uNTx/uCr were associated; at day 10, uNTx/uCr and CSC had a Spearman rank correlation coefficient of 0.62 (P < .01).
Relief of symptoms
Ascribing of symptoms to hypercalcemia was per the treating physician's assessment and captured on prespecified hypercalcemia symptom case report forms. Of the 19 patients with at least 1 symptom attributed to hypercalcemia, 10 (53%) reported improvement or resolution of at least 1 symptom; for 8 of these patients (42%), improvement or resolution occurred by day 10. For 6 patients (32%), all symptoms were improved or resolved; resolution occurred by day 10 for 4 of these patients (21%). Of the 48 symptoms reported at baseline, 15 (31%) were resolved, 5 (10%) improved, 26 (54%) remained stable, and 2 (4%) became worse. Among the 9 patients with symptoms of cognitive impairment, 5 (56%) had resolution of at least 1 cognitive symptom and 4 (44%) had resolution of all cognitive symptoms by day 10. At the symptom level, of the 12 cognitive symptoms reported, 7 (58%) resolved, 4 (33%) remained stable, and 1 (8%) worsened. At day 10, the 6 patients for whom all symptoms improved or resolved had a lower mean (SD) CSC (10.7 [1.0] mg/dL) than the 8 patients whose symptoms remained stable (12.2 [2.1] mg/dL). Also at day 10, the 4 patients for whom all cognitive symptoms improved or resolved had a mean (SD) CSC of 10.4 (1.1) mg/dL, compared with 13.3 (3.2) mg/dL for the 3 patients for whom all cognitive symptoms remained stable. The subject incidence of HCM symptom improvement, as reflected in CTCAE grade decrease from baseline up to study day 10 by visit, was summarized by individual HCM symptoms as well as by overall (ie, any and all) HCM symptoms.
Other laboratory results
Median PTHrP decreased only slightly from 4.2 (range, 0.5–24) pmol/L at baseline to 3.7 (range, 0.2–29) pmol/L at day 29 and remained at approximately that level through day 57. Median uNTx/uCr decreased from 76.9 (range, 7–425) nmol/mmol at baseline to 19.9 (range, 7–87) nmol/mmol at day 10 and remained close to that level until day 57. Two patients had isolated nonsymptomatic episodes of CSC levels ≤8.0 mg/dL (CTCAE grade 2 hypocalcemia); 1 of these was reported as a treatment-related adverse event. No concurrent factors (eg, CrCl, PTHrP, or number of denosumab doses) were identified that might have contributed to low calcium levels. Median CrCl values did not change significantly from baseline to day 57. No CTCAE grade 3 hypocalcemia (CSC <7.0 mg/dL) was reported in any patient. No unexpected or clinically meaningful changes were observed in other laboratory values analyzed.
Safety results
As expected in this population of patients with advanced cancer, 32 patients (97%) experienced at least 1 adverse event during the study (Table 3). The most frequently reported adverse events were nausea (10 patients, 30%); dyspnea, headache, peripheral edema, and vomiting (8 patients, 24% each); and constipation, decreased appetite, and diarrhea (7 patients, 21% each). Thirteen patients (39%) experienced adverse events that were considered by the investigator to be potentially treatment-related; the most frequently reported were hypophosphatemia and nausea, each reported in 4 patients (12%). Serious adverse events were reported in 29 patients (88%). The most frequently reported serious adverse events, none considered treatment-related, were cancer progression (8 patients, 24%). Five patients experienced worsening of hypercalcemia (15%), and 6 patients (18%) experienced serious gastrointestinal events, including diarrhea (2 patients, 6%) and colitis, nausea, esophageal obstruction, and vomiting (1 patient, 3%, each). Two serious adverse events were considered possibly treatment-related by the investigator: cardiac arrest and colitis, each in 1 patient. No osteonecrosis of the jaw was reported. At the time of the data cutoff, 28 patients (85%) had died. The estimated median survival time was 71 days (95% CI, 41–158).
Table 3.
Patients Reporting Adverse Eventsa
| Denosumab Treatment Group, N = 33 | |
|---|---|
| Adverse events, all | 32 (97) |
| Serious adverse events | 29 (88) |
| Serious adverse events reported in ≥2 patients | |
| Cancer progressionb | 8 (24) |
| Hypercalcemia | 5 (15) |
| Dyspnea | 3 (9) |
| Diarrhea | 2 (6) |
| Tachycardia | 2 (6) |
| Adverse events resulting in study discontinuationc | 8 (24) |
| Adverse events reported in ≥20% of patients | |
| Nausea | 10 (30) |
| Dyspnea | 8 (24) |
| Headache | 8 (24) |
| Peripheral edema | 8 (24) |
| Vomiting | 8 (24) |
| Constipation | 7 (21) |
| Decreased appetite | 7 (21) |
| Diarrhea | 7 (21) |
Results are shown as n (%).
All serious adverse events of cancer progression were considered related to the underlying disease and not to denosumab treatment. Group includes head and neck, lung neoplasm malignant, non-small-cell lung cancers, and multiple myeloma.
Adverse events resulting in discontinuation from the study included progression of breast cancer in 2 patients; progression of bladder cancer, head and neck cancer, non-small-cell lung cancer, or renal cell carcinoma (in 1 patient each); and hepatic failure and hypoxia (in 1 patient each).
Discussion
In this study of denosumab in patients with recurrent or refractory hypercalcemia despite recent treatment with iv bisphosphonates, 64% of patients responded to treatment by day 10 and 70% responded over the course of the study. Response occurred in a median of 9 days and had an estimated median duration of 104 days. Twenty-one patients (64%) had a complete response over the course of the study, with CSC levels ≤10.8 mg/dL, with a median duration of 34 days. For 12 (36%) of these patients, this complete response occurred by day 10. The response rate, the complete response rate, and the prolonged duration of response observed in this study are noteworthy because the enrolled patients had not responded or had relapsed despite bisphosphonate treatment within a median of 17 days before study entry. Typically, such patients would have few therapeutic options available to them. Similar proportions of patients responded to treatment regardless of baseline factors including age, racial group, bone metastases, hypercalcemia symptoms, baseline levels of PTHrP, or renal impairment. Denosumab also relieved symptoms of hypercalcemia, including cognitive symptoms, for many patients. Because the development of HCM is a poor prognostic indicator in the cancer population with high mortality, palliation of symptoms is essential and clinically meaningful for patients. We note that the efficacy data presented in this report are of similar but generally lesser magnitude than the interim data we published previously for the first 15 patients (20). For example, response rate by day 10 was 80% at the interim analysis and was 64% in this primary analysis. The larger numbers of patients included in this report should be more representative of the anticipated population effect, and regardless, the reported results of both analyses are clinically meaningful.
We acknowledge several limitations of this study. The study included a relatively small sample size and short duration of follow-up, and long-term data for this population were not available at the time of the primary analysis. However, given the mortality rate of 85% at the primary analysis point of this study, the short follow-up duration reflects the anticipated life expectancy in these patients. Also, although the study was appropriately powered to evaluate the primary objective of estimating the proportion of subjects who respond within 10 days of denosumab (defined as CSC ≤11.5 mg/dL), the prespecified subgroups analyses were descriptive and not powered to detect statistical differences. Evaluation of symptoms is confounded by the fact that many of the reported symptoms can be present in patients with advanced cancer without hypercalcemia.
However, this study also has significant strengths. The study was international in scope, had no upper age limit, and enrolled patients who had a variety of solid tumors and hematologic malignancies. No renal toxicity was observed, and patients with impaired renal function responded to denosumab therapy. Because denosumab is neither metabolized by nor excreted from the kidneys, it can be administered to patients with advanced renal failure, a limitation with iv bisphosphonate administration. Denosumab is administered sc, obviating the need for inpatient admission, use of infusion facilities, or iv access. No agent has been approved for the treatment of HCM since the 2001 approval of zoledronic acid. Albeit investigational in this treatment setting, denosumab has the potential to be a new effective agent for this difficult-to-treat patient population.
In conclusion, in this study in patients with HCM despite recent treatment with iv bisphosphonates, denosumab effectively lowered serum calcium in a majority of patients and induced durable responses beyond those seen with prior therapies in patients enrolled in this study.
Acknowledgments
We acknowledge the medical writing assistance of Albert Y. Rhee of Amgen Inc and Susan E. Hudson, a medical writing contractor employed by Amgen. Their written permission has been obtained.
This study was sponsored by Amgen Inc, which contributed to the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, and approval of the manuscript; and the decision to submit the manuscript for publication. This work was supported by Amgen Inc. This study was supported by the National Institutes of Health/National Cancer Institute under award number P30CA01662.
All authors had full access to all data and participated in the writing and editing of the manuscript. M.I.H. and R.K.J. take responsibility for the integrity of the data and the accuracy of the data analysis. W.Y. of Amgen Inc performed the statistical analyses of the data in this manuscript.
Data included in this submission were presented in part as a poster at the European Cancer Congress, Amsterdam, Netherlands, September 27 to October 1, 2013, and as an oral poster presentation at the American Society for Bone and Mineral Research Congress, Baltimore, MD, October 4–7, 2013. Abstracts of some data from this study have also been accepted for poster presentations at the American Society of Nephrology, Atlanta, GA, November 5–10, 2013; the 13th International Congress on Cancer-Induced Bone Disease, Miami, FL, November 6–9, 2013; and the American Society of Hematology, New Orleans, LA, December 7–10, 2013.
Disclosure Summary: M.I.H., I.G., and W.M. received research grants from Amgen Inc. P.Z. owns stock and/or stock options in CytRx Inc, Galena Biopharma, and RxII. D.T. received research grants from Novartis. W.Y., A.B., and R.J. are employees of Amgen Inc and have received stock and/or stock options in Amgen Inc. K.I., R.G., S.L., B.Y., A.B., A.J., and G.T. have nothing to disclose.
Footnotes
- CI
- confidence interval
- CrCl
- creatinine clearance
- CSC
- corrected serum calcium
- CTCAE
- Common Terminology Criteria for Adverse Events
- HCM
- hypercalcemia of malignancy
- Q1
- Q3, 25th and 75th percentiles
- RANKL
- receptor activator of nuclear factor-κB ligand
- ULN
- upper limit of normal
- uNTx/uCr
- urinary-N-telopeptide corrected for urinary creatinine.
References
- 1. Rosner MH, Dalkin AC. Onco-nephrology: the pathophysiology and treatment of malignancy-associated hypercalcemia. Clin J Am Soc Nephrol. 2012;7:1722–1729 [DOI] [PubMed] [Google Scholar]
- 2. Stewart AF. Clinical practice. Hypercalcemia associated with cancer. N Engl J Med. 2005;352:373–379 [DOI] [PubMed] [Google Scholar]
- 3. Gastanaga V, Jain R, Pirolli M, et al. Prevalence of hypercalcemia of malignancy in the United States: projection methods using oncology electronic health records (EHR). Eur J Cancer. 2013;49:S1–S1028, Abstract 1415, p. 25 [Google Scholar]
- 4. Ralston SH, Gallacher SJ, Patel U, Campbell J, Boyle IT. Cancer-associated hypercalcemia: morbidity and mortality. Clinical experience in 126 treated patients. Ann Intern Med. 1990;112:499–504 [DOI] [PubMed] [Google Scholar]
- 5. Burtis WJ, Wu T, Bunch C, et al. Identification of a novel 17,000-dalton parathyroid hormone-like adenylate cyclase-stimulating protein from a tumor associated with humoral hypercalcemia of malignancy. J Biol Chem. 1987;262:7151–7156 [PubMed] [Google Scholar]
- 6. Moseley JM, Kubota M, Diefenbach-Jagger H, et al. Parathyroid hormone-related protein purified from a human lung cancer cell line. Proc Natl Acad Sci U S A. 1987;84:5048–5052 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Strewler GJ, Stern PH, Jacobs JW, et al. Parathyroid hormonelike protein from human renal carcinoma cells. Structural and functional homology with parathyroid hormone. J Clin Invest. 1987;80:1803–1807 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Wysolmerski JJ, Broadus AE. Hypercalcemia of malignancy: the central role of parathyroid hormone-related protein. Annu Rev Med. 1994;45:189–200 [DOI] [PubMed] [Google Scholar]
- 9. LeGrand SB, Leskuski D, Zama I. Narrative review: furosemide for hypercalcemia: an unproven yet common practice. Ann Intern Med. 2008;149:259–263 [DOI] [PubMed] [Google Scholar]
- 10. Major P, Lortholary A, Hon J, et al. Zoledronic acid is superior to pamidronate in the treatment of hypercalcemia of malignancy: a pooled analysis of two randomized, controlled clinical trials. J Clin Oncol. 2001;19:558–567 [DOI] [PubMed] [Google Scholar]
- 11. Major PP, Coleman RE. Zoledronic acid in the treatment of hypercalcemia of malignancy: results of the international clinical development program. Semin Oncol. 2001;28:17–24 [DOI] [PubMed] [Google Scholar]
- 12. Aredia (pamidronate disodium) [package insert]. East Hanover, NJ: Novartis Pharmaceuticals Corporation; 2012 [Google Scholar]
- 13. Zometa (zoledronic acid) [package insert]. East Hanover, NJ: Novartis Pharmaceuticals Corporation; 2012 [Google Scholar]
- 14. Morony S, Warmington K, Adamu S, et al. The inhibition of RANKL causes greater suppression of bone resorption and hypercalcemia compared with bisphosphonates in two models of humoral hypercalcemia of malignancy. Endocrinology. 2005;146:3235–3243 [DOI] [PubMed] [Google Scholar]
- 15. Oyajobi BO, Anderson DM, Traianedes K, Williams PJ, Yoneda T, Mundy GR. Therapeutic efficacy of a soluble receptor activator of nuclear factor kappaB-IgG Fc fusion protein in suppressing bone resorption and hypercalcemia in a model of humoral hypercalcemia of malignancy. Cancer Res. 2001;61:2572–2578 [PubMed] [Google Scholar]
- 16. Fizazi K, Carducci M, Smith M, et al. Denosumab versus zoledronic acid for treatment of bone metastases in men with castration-resistant prostate cancer: a randomised, double-blind study. Lancet. 2011;377:813–822 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Henry DH, Costa L, Goldwasser F, et al. Randomized, double-blind study of denosumab versus zoledronic acid in the treatment of bone metastases in patients with advanced cancer (excluding breast and prostate cancer) or multiple myeloma. J Clin Oncol. 2011;29:1125–1132 [DOI] [PubMed] [Google Scholar]
- 18. Stopeck AT, Lipton A, Body JJ, et al. Denosumab compared with zoledronic acid for the treatment of bone metastases in patients with advanced breast cancer: a randomized, double-blind study. J Clin Oncol. 2010;28:5132–5139 [DOI] [PubMed] [Google Scholar]
- 19. Diel IJ, Body JJ, Stopeck AT, et al. Effect of denosumab treatment on prevention of hypercalcemia of malignancy in cancer patients with metastatic bone disease. Eur J Cancer. 2011; 47:S237, Abstract 3051 [DOI] [PubMed] [Google Scholar]
- 20. Hu MI, Glezerman I, Leboulleux S, et al. Denosumab for patients with persistent or relapsed hypercalcemia of malignancy despite recent bisphosphonate treatment. J Natl Cancer Inst. 2013;105:1417–1420 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Cancer Therapy Evaluation Program. Common Terminology Criteria for Adverse Events (CTCAE), Version 3.0. http://ctep.cancer.gov/protocoldevelopment/electronic_applications/docs/ctcaev3.pdf Accessed August 8, 2013
- 22. Thomas D, Henshaw R, Skubitz K, et al. Denosumab in patients with giant-cell tumour of bone: an open-label, phase 2 study. Lancet Oncol. 2010;11:275–280 [DOI] [PubMed] [Google Scholar]
- 23. XGEVA (denosumab solution for injection). Product monograph. Amgen Canada Inc; 2013 [Google Scholar]
- 24. Chawla S, Henshaw R, Seeger L, et al. Safety and efficacy of denosumab for adults and skeletally mature adolescents with giant cell tumour of bone: interim analysis of an open-label, parallel-group, phase 2 study. Lancet Oncol. 2013;14:901–908 [DOI] [PubMed] [Google Scholar]