Denosumab treatment of prostate cancer with bone metastases and increased urine N-telopeptide levels after therapy with intravenous bisphosphonates: results of a randomized phase II trial.

Abstract

Patients with bone metastases have high rates of RANKL driven bone resorption and an increased risk of skeletal morbidity. Osteoclast mediated bone resorption can be assessed by measuring urine N-telopeptide and can be inhibited by denosumab, a fully human antibody against RANKL.
Eligible patients (111) had bone metastases from prostate cancer, other solid tumors or multiple myeloma, 1 or more bone lesions and urine N-telopeptide greater than 50 nM bone collagen equivalents per mM creatinine (urine N-telopeptide greater than 50) despite the use of intravenous bisphosphonates. Patients were stratified by cancer type and screening urine N-telopeptide, and randomized to continue intravenous bisphosphonates every 4 weeks or receive 180 mg subcutaneous denosumab every 4 weeks or 180 mg every 12 weeks. The primary end point was the proportion of patients with urine N-telopeptide less than 50 at week 13. We report the efficacy results for the subset of patients with prostate cancer.
Patients with prostate cancer represented 45% (50 of 111) of the study population. At week 13, 22 of 32 (69%) patients in the denosumab arms had urine N-telopeptide less than 50 vs 3 of 16 (19%) in the intravenous bisphosphonates cohort. At week 25, 22 of 32 (69%) denosumab treated patients continued to have urine N-telopeptide less than 50 vs 5 of 16 (31%) treated with intravenous bisphosphonates. Grade 4, asymptomatic, reversible hypophosphatemia, possibly related to denosumab, was reported in 1 patient.
In patients with prostate cancer related bone metastases and increased urine N-telopeptide despite intravenous bisphosphonate treatment, denosumab normalized urine N-telopeptide levels more frequently than ongoing intravenous bisphosphonates.

Full text

Page 1

Denosumab Treatment of Prostate Cancer With Bone Metastases and
Increased Urine N-Telopeptide Levels After Therapy With Intravenous
Bisphosphonates: Results of a Randomized Phase II Trial
Karim Fizazi,*,† Linda Bosserman,‡ Guozhi Gao,§ Tomas Skacel§ and
Richard Markus§
From the Institut Gustave Roussy and University of Paris XI, Villejuif, France (KF), Wilshire Oncology Medical Group, La Verne (LB),
Amgen Inc., San Francisco (GG) and Amgen Inc., Thousand Oaks (RM), California, and Amgen (Europe) GmbH, Zug,
Switzerland (TS)
Purpose: Patients with bone metastases have high rates of RANKL driven bone
resorption and an increased risk of skeletal morbidity. Osteoclast mediated bone
resorption can be assessed by measuring urine N-telopeptide and can be inhibited by
denosumab, a fully human antibody against RANKL.
Materials and Methods: Eligible patients (111) had bone metastases from prostate
cancer, other solid tumors or multiple myeloma, 1 or more bone lesions and urine
N-telopeptide greater than 50 nM bone collagen equivalents per mM creatinine
(urine N-telopeptide greater than 50) despite the use of intravenous bisphospho-
nates. Patients were stratified by cancer type and screening urine N-telopeptide, and
randomized to continue intravenous bisphosphonates every 4 weeks or receive 180
mg subcutaneous denosumab every 4 weeks or 180 mg every 12 weeks. The primary
end point was the proportion of patients with urine N-telopeptide less than 50 at
week 13. We report the efficacy results for the subset of patients with prostate
cancer.
Results: Patients with prostate cancer represented 45% (50 of 111) of the study
population. At week 13, 22 of 32 (69%) patients in the denosumab arms had urine
N-telopeptide less than 50 vs 3 of 16 (19%) in the intravenous bisphosphonates
cohort. At week 25, 22 of 32 (69%) denosumab treated patients continued to have
urine N-telopeptide less than 50 vs 5 of 16 (31%) treated with intravenous bisphos-
phonates. Grade 4, asymptomatic, reversible hypophosphatemia, possibly related to
denosumab, was reported in 1 patient.
Conclusions: In patients with prostate cancer related bone metastases and in-
creased urine N-telopeptide despite intravenous bisphosphonate treatment, deno-
sumab normalized urine N-telopeptide levels more frequently than ongoing intrave-
nous bisphosphonates.
Key Words: denosumab; osteoclasts; prostatic neoplasms; RANK ligand;
urinary N-telopeptide of type I collagen, human
PROSTATE cancer was diagnosed in
more than a half-million men world-
wide in 2002 and was the cause of
death for more than a quarter-million
men.1In up to 75% of patients with
advanced disease prostate cancer me-
tastasizes to bone, which may result
in skeletal complications including
pathological skeletal fractures, spinal
cord compression, or radiation ther-
apy or surgery to bone, collectively
known as skeletal related events.2
Abbreviations
and Acronyms
AE ? adverse event
BCE ? bone collagen equivalents
BP ? bisphosphonates
Cr ? creatinine
ECOG ? Eastern Cooperative
Oncology Group
IV ? intravenous
Q4W ? every 4 weeks
Q12W ? every 12 weeks
RANKL ? receptor activator of
nuclear factor ?B ligand
SC ? subcutaneous
SRE ? skeletal related event
uNTx ? urine N-telopeptide
Submitted for publication December 22, 2008.
Study received institutional review board/
ethics committee approval.
Supported by Amgen Inc.
Clinical Trial Registration NCT00104650 (www.
clinicaltrials.gov).
* Correspondence: Department of Medicine,
Institut Gustave Roussy and University of Paris XI,
39 Rue Camille Desmoulins, 94800 Villejuif,
France (telephone: 33 1 42 11 43 17; FAX: 33 1 42
11 52 11; e-mail: fizazi@igr.fr).
† Financial interest and/or other relationship
with Amgen, Novartis, AstraZeneca, Sanofi-
Aventis, Ipsen-Beaufour, Pharmion, Bristol Myers
Squibb and Takeda.
‡ Financial interest and/or other relationship
with Amgen and Pfizer.
§ Financial interest and/or other relationship
with Amgen.
0022-5347/09/1822-0509/0
THE JOURNAL OF UROLOGY®
Copyright © 2009 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 182, 509-516, August 2009
Printed in U.S.A.
DOI:10.1016/j.juro.2009.04.023
www.jurology.com
509

Page 2

In patients with bone metastases tumor cells in
the bone disrupt the local balance between bone
resorption (mediated by osteoclasts) and bone for-
mation (mediated by osteoblasts) in the affected re-
gion.3The result is increased, unbalanced bone
turnover that contributes to a loss of structural in-
tegrity in the bone. This process creates a vicious
cycle of tumor growth and bone destruction at sites
of metastatic cells. Within the bone microenviron-
ment growth factors secreted by tumor cells induce
stromal cells and osteoblasts to secrete the receptor
activator of NF-?B ligand (RANKL), an essential
mediator of osteoclast formation, function and sur-
vival.4Binding of RANKL to its receptor, RANK, on
osteoclasts and osteoclast precursors promotes in-
creased osteoclast formation and bone resorption.
Growth factors released from the bone matrix promote
furthertumorgrowthandperpetuatetheviciouscycle.
The key role of the RANKL/osteoprotegerin axis in the
cycle of bone metastases from prostate cancer was
shown in in vitro models.5The favorable effect of
RANKL inhibition was subsequently demonstrated
in animal models of bone metastases.6
Although prostate cancer related bone metasta-
ses often appear osteoblastic on radiographs, osteol-
ysis and poor mineralization can be seen microscop-
ically.5,7,8Prostate cancer bone metastases are
characterized by increased deposition of quickly
formed, immature, woven bone,9which is more brit-
tle and weaker than normal bone tissue,10potentially
increasing the risk of fracture and other skeletal com-
plications. Cytotoxic, glucocorticoid and androgen de-
privation therapies for cancer may also decrease bone
mineral density, accelerating bone loss and increasing
fracture risk.11
Markers of bone turnover such as urine N-te-
lopeptide, used in research, are indicators of clinical
prognosis in patients with prostate cancer related
bone metastases. In these patients increased levels
of uNTx (greater than 50 nM BCE/mM creatinine
[uNTx greater than 50]) represent excess bone re-
sorption, and are associated with increased risk of
SREs, disease progression and death.12
Current treatments for reducing skeletal compli-
cations of bone metastases in prostate cancer in-
clude localized radiation to bone and the intrave-
nous bisphosphonate zoledronic acid.13Other IV
BPs (pamidronate and ibandronate), which are ef-
fective in bone metastases with other tumor types,
failed to show significant activity in prostate cancer
in randomized trials. More effective, safer treat-
ments that are convenient and well tolerated are
needed.
Denosumab is an investigational fully human
monoclonal antibody, administered subcutaneously,
that inhibits osteoclast mediated bone resorption in
bone metastases from solid tumors and multiple
myeloma.14We recently reported results from a
phase II, randomized, active controlled, open label
trial that evaluated the effects of denosumab on
bone turnover markers and SREs in patients with
bone metastases and increased uNTx levels despite
previous treatment with zoledronic acid.15This re-
port focuses on the subset of patients with prostate
cancer from that study, describing the effects of de-
nosumab on bone resorption markers, the incidence
of skeletal complications and safety in these pa-
tients.
MATERIALS AND METHODS
Patients
Eligible patients were 18 years old or older, with prostate
cancer, other solid carcinomas (except lung cancer) or
multiple myeloma with radiographic evidence of 1 or more
bone lesions and an ECOG performance status of 2 or less.
All patients had received 8 or more weeks of treatment
with IV BP for treatment of bone metastases, yet contin-
ued to have evidence of excessive bone resorption (uNTx
levels greater than 50). Patients were excluded from study
if they had more than 2 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 study was approved by the
institutional review board or ethics committee for each
site. All patients provided written informed consent.
Study Design and Procedures
This multicenter, phase II, randomized, open label, active
controlled study, conducted at 26 centers in Europe and
North America from December 2, 2004 through January
20, 2008, compared 2 different doses of subcutaneous de-
nosumab and IV BP. Patients were randomly assigned in
a 1:1:1 ratio to continue IV BP therapy every 4 weeks or to
discontinue IV BP therapy and receive subcutaneous in-
jections of denosumab 180 mg Q4W or Q12W (fig. 1).
Randomization was stratified by cancer type (prostate,
breast, or other solid tumors and multiple myeloma) and
baseline uNTx (50 to 100 or greater than 100). All patients
were instructed to take daily supplements of calcium (500
mg) and vitamin D (400 or more IU). The treatment period
was 25 weeks, after which patients were eligible to enroll
in a separate 2-year extension study. Those who did not
participate in the extension study were followed in post-
treatment visits at weeks 33, 45 and 57. We describe final
results of the 25-week treatment phase of the study and 32
weeks of off treatment followup (a total of up to 57 weeks)
for the patients with prostate cancer (50) at study entry.
This subset of patients with prostate cancer had prior
treatment with zoledronic acid. Most patients with pros-
tate cancer had evidence of bone metastases despite ongo-
ing androgen deprivation therapy/antiandrogens (78%) so
that the disease was considered castration resistant in the
majority of patients.
Initial and periodic study assessments included a med-
ical history and physical examination, vital signs, electro-
cardiograms, radiographic evaluations of the spine, and
DENOSUMAB FOR PROSTATE CANCER BONE METASTASES
510

Page 3

laboratory assessments including hematology, serum
chemistry and bone resorption markers. Prostate specific
antigen levels were not collected in this study. Concurrent
antineoplastic or hormonal therapy was permitted during
study treatment at investigator discretion as long as no
changes in regimens or agents were planned within 4
weeks of randomization.
End Points
Efficacy end points included changes in bone resorption
markers and the incidence of SREs. The primary efficacy
end point was the proportion of patients with uNTx less
than 50 at week 13. Secondary end points included the
proportion of patients achieving uNTx less than 50 at
week 25, time to reduction of uNTx to less than 50 and
duration of uNTx level less than 50. The study also eval-
uated the patient incidence of SREs, defined as a patho-
logical bone fracture, spinal cord compression, or surgery
or radiation therapy to bone; the time from enrollment to
first on-study SRE; and the incidence of hypercalcemia.
Safety end points were assessed at regular intervals, and
included AEs graded using the Common Terminology Cri-
teria for Adverse Events version 3, changes from baseline
in laboratory assessments and the incidence of anti-deno-
sumab antibody formation.
Statistical Analysis
The statistical analysis plan for this study has been de-
scribed previously.15The 2 denosumab arms were pooled
for analysis. The primary analysis was conducted on all
patients who were randomized, received at least 1 dose of
investigational product, and had uNTx measurements at
baseline and at least 1 other time after baseline. Safety
analyses were performed on randomized patients who re-
ceived at least 1 dose of investigational drug.
RESULTS
Patients
Of 111 enrolled patients 50 (45%) had prostate can-
cer at baseline. The median time to randomization
from the diagnosis of bone metastases was 0.6 years,
but ranged up to 4 years (table 1). Other patients
enrolled in the study had breast cancer (46 of 111,
41%), multiple myeloma (9 of 111, 8%) and other
solid tumors (6 of 111, 5%). One patient with pros-
tate cancer assigned to the IV BP group was ineli-
gible after enrollment while 49 patients received
treatment (fig. 1). In the prostate cancer subset 27 of
33 patients (82%) in the denosumab groups and 10 of
17 (59%) in the IV BP group completed the full 25
weeks of treatment. The most common reasons for
study discontinuation were death from underlying
disease and disease progression.
While baseline demographics and disease charac-
teristics were generally balanced among treatment
groups, more patients in the denosumab groups
than the IV BP group (33% vs 24%) had an ECOG
status of 2. The median age of patients with prostate
cancer was 68 years. At baseline the extent of bone
resorption was greater in more patients randomized
to denosumab compared to those randomized to IV
BP, as indicated by uNTx levels (table 1). All pa-
tients had prior treatment with zoledronic acid with
a median of 6 months treatment before randomiza-
tion (IV BP 4 months, denosumab 7 months). No
other IV BPs were used by patients in the prostate
cancer subset. Approximately half (17 of 33, 52%) of
Randomized
N = 111
SC Denosumab
180 mg Q4W
n = 17
SC Denosumab
180 mg Q12W
n = 16
IV Bisphosphonates
Q4W
n = 17
Received
IV Bisphosphonates
n = 16
Ineligible
n = 1
Received SC
Denosumab
180 mg Q4W
n = 17
Received SC
Denosumab
180 mg Q12W
n = 16
Prostate Cancer
n = 50
Other Cancers
n = 61
Figure 1. Random treatment assignments
DENOSUMAB FOR PROSTATE CANCER BONE METASTASES
511

Page 4

the patients randomized to the denosumab groups
and a quarter (4 of 17, 24%) of those randomized to
the IV BP group experienced SREs before entering
the study (a strong predictor of subsequent SREs).16
Efficacy
Bone resorption. After 13 weeks suppression of
bone resorption was observed in a greater propor-
tion of patients in the denosumab groups than in the
IV BP group as demonstrated by the proportion of
patients with uNTx levels less than 50 (22 of 32, 69%
vs 3 of 16, 19%; primary end point, fig. 2, A) and the
median percentage reduction in uNTx levels from
baseline (84% vs 32%, fig. 2, B). Denosumab induced
reduction in bone resorption was sustained at week
25, with 22 of 32 (69%) patients in the denosumab
groups and 5 of 16 (31%) in the IV BP group con-
tinuing to have uNTx less than 50 (fig. 2, C).
Denosumab induced suppression of bone resorp-
tion was rapid, observed as early as 2 weeks (the
first point at which uNTx was evaluated on-study,
fig. 2, B). Median time to reduction of uNTx less
than 50 was 10 days in the denosumab arms and 88
days in IV BP treated patients (stratified Cox model
p ?0.001). Normalization of uNTx levels occurred
more frequently in the pooled denosumab groups
than the IV BP group at weeks 13 and 25 regardless
of screening uNTx levels (fig. 2, A and C). The effects
of denosumab were sustained throughout the treat-
ment period. For most patients receiving deno-
sumab, uNTx never increased above less than 50
nm/mmol after suppression. The estimated 25th
percentile of duration of maintaining uNTx less
than 50 was 160 days (95% CI 56–,) in the deno-
sumab groups and only 28 days (95% CI 21–78) in
the IV BP group (p ? 0.016). The 25th percentile
was evaluated for this comparison because 50% (me-
dian) of the patients in the denosumab group never
had their uNTx levels revert to a level above 50 nm.
Skeletal complications. A lower proportion of pa-
tients with prostate cancer in the denosumab groups
than in the IV BP group experienced an on-study
SRE (1 of 33 [3%] treated with denosumab and 3 of
16 [19%] treated with IV BP). This difference was
similar to that seen in the overall study population
(fig. 3).15No cases of hypercalcemia were reported.
Safety
Safety results were as expected in a population with
advanced cancer receiving systemic chemotherapy
and androgen deprivation therapy. The incidence
and types of AEs were similar in both treatment
groups and similar to those for the entire population
(table 2). In the prostate cancer subset, AEs were
reported in 31 of 33 patients (94%) in the pooled
denosumab group and 16 of 16 patients (100%) in
the IV BP group. AEs considered potentially treat-
ment related were reported in 9 of 33 patients (27%)
receiving denosumab and 2 of 16 patients (12%)
receiving IV BP. Similar low rates of serious AEs
Table 1. Baseline demographics and disease characteristics in prostate cancer subset
IV BP Q4W
Denosumab
Q12W Q4WAll
No. men (%)
Mean ? SD age
No. ECOG status (%):
0 or 1
2
Median yrs since original diagnosis (min, max)
No. treatment received (%):
Chemotherapy
Androgen deprivation/antiandrogens
Median mos on BP (min, max)
Median yrs since bone metastases diagnosis (min, max)
No. bone metastases (%):
2 or Less
Greater than 2
No. previous SREs (%):
0
1 or More
Median ml/min/1.73 m2estimated glomerular filtration
rate (min, max)
Median nM BCE mM/Cr baseline uNTx corrected for Cr
(Q1, Q3)
No. less than 50 nM BCE mM/Cr (%)
No. 50–100 nM BCE mM/Cr (%)
No. greater than 100 nM BCE mM/Cr (%)
17
69.5 ? 9.5
(100) 16
67.4 ? 12.3
(100.0)17
64.5 ? 8.7
(100.0)33
65.9 ? 10.5
(100.0)
13(76)
(24)
11 (69)
(31)
11(65)
(35)
22
11
(67)
(33) 4
2.6
5
3.8
6
3.8(0.2, 13.2)(0.1, 8.7)(0.3, 11.1)3.8(0.1, 11.1)
7(41)
(82)
5(31)
(69)
10
14
(59)
(82)
15
25
(46)
(76) 1411
3.7
0.5
(1.6, 28.7)
(0.1, 2.7)
6.8
0.6
(2.1, 29.1)
(0.0, 4.1)
6.60
0.7
(1.8, 30.2)
(0.1, 4.1)
6.60
0.6
(1.8, 30.2)
(0.0, 4.1)
0(0.0)1 (6.3)
(93.8)
1(5.9)
(94.1)
2(6.1)
(93.9)17(100.0)151631
13(76)
(24)
8
8
(50.0)
(50.0)
8
9
(47)
(53)
16
17
86
(48)
(52)4
102(49, 183) 97(39, 141) 81(38, 164)(38, 164)
103.4 (62.5, 212.5)149.3 (48.8, 346.8)129.9 (64.6, 278.7)129.9 (50.8, 285.6)
2
5
(12)
(29)
(59)
5
1
(31)
(6)
(63)
3
2
(18)
(12)
(71)
8
3
(24)
(9)
(67)10 1012 22
DENOSUMAB FOR PROSTATE CANCER BONE METASTASES
512

Page 5

considered potentially treatment related were re-
ported in both treatment groups.
Expected, slight decreases in serum calcium lev-
els were observed based on laboratory values (me-
dian change [Q1, Q3]) at study week 2 after treat-
ment with IV BP (?0.13 mM/l [?0.21, ?0.02]) or
denosumab (?0.23 mM/l [?0.30, ?0.07]). Serum cal-
cium levels remained relatively unchanged com-
pared with baseline levels in the IV BP arm and the
pooled denosumab arm through week 25. Median
(Q1, Q3) changes in alkaline phosphatase levels
were comparable between the IV BP arm and the
denosumab arm after week 2 (?12.5 [?59.0, 39.5]
and ?14.0 [?84.0, 18.0], respectively) and after
week 25 (?167.0 [?325.0, 110.0] and ?72.0 [?211.5,
?38.0], respectively).
Only 1 AE of grade 3 or greater that was consid-
ered possibly treatment related was reported among
the patients with prostate cancer. This patient re-
ceived 180 mg denosumab Q4W and experienced
asymptomatic transient hypophosphatemia (grade
4), which accompanied rapid underlying cancer pro-
gression and biological evidence of hyperparathy-
roidism (resolved after 5 weeks).
Two patients in the prostate cancer subset expe-
rienced grade 3 hypocalcemia. In 1 patient treated
with denosumab Q4W the hypocalcemia was consid-
ered serious but not treatment related. Another pa-
tient received IV BP and also experienced grade 3
hypocalcemia that was not considered serious but
was deemed treatment related by the investigator.
Expected, transient decreases in serum calcium lev-
els were observed in both treatment groups. The
rates and nature of deaths on the study were as
expected, and none were considered treatment re-
lated. Denosumab treatment had no marked effect
on renal or hepatic function and no patient experi-
enced neutralizing anti-denosumab antibodies or os-
teonecrosis of the jaw.
DISCUSSION
The population described in this study may repre-
sent the most difficult patients in whom to treat
bone metastases and to normalize bone resorption.
They had poor prognoses as well as an increased risk
of SREs and death as indicated by increased levels of
uNTx despite at least 8 weeks of IV BP treatment
before study entry. In these patients regardless of
baseline uNTx levels denosumab induced suppres-
sion of bone resorption occurred rapidly and contin-
ued during the course of treatment. The incidence of
SREs in the prostate cancer subgroups was simi-
lar to the SRE rate in the overall population and
lower in the denosumab treated group (3%) than
in the IV BP group (19%), although the small
population size does not allow a statistical com-
parison. The adverse event profiles in both groups
were similar and denosumab appeared to be well
tolerated overall.
0
20
40
60
80
100
IV BP
Denosumab
IV BP
Denosumab
Proportion of patients (%, 95% CI) with
uNTx/Cr < 50 nM BCE/mM at 13 weeks
N = 16 N = 32
All patients with
prostate cancer
N = 6 N = 12
50–100 nm
BCE/mM
N = 10 N = 20
>100 nM
BCE/mM
Denosumab 180 mg Q12W
Denosumab 180 mg Q4W
IV BPs
Total Denosumab
-100
-80
-60
-40
-20
0
20
40
60
259 13172125
Median (Q1, Q3) change in uNTx/Cr (%)
(Prostate cancer patients)
Visit Week
Baseline
0
20
40
60
80
100
All patients with
prostate cancer
50–100 nm
BCE/mM
>100 nM
BCE/mM
Proportion of patients (%, 95% CI) with
uNTx/Cr < 50 nM BCE/mM at 25 weeks
N = 16
uNTx/Cr at Baseline
uNTx/Cr at Baseline
N = 32N = 6 N = 12 N = 10 N = 20
A
B
C
Figure 2. Proportion of patients with prostate cancer achieving uNTx less than 50 nM BCE/mM creatinine at 13 and 25 weeks for all
patients and stratified by baseline uNTx (A and C), and median (Q1, Q3) percent change in uNTx/Cr over time (B).
0
2
4
6
8
10
12
14
16
18
20
IV BP
N = 35
Denosumab
180 mg Q4W
N = 38
Denosumab
180 mg Q12W
N = 35
Proportion of patients (%)
experiencing SREs
Denosumab
All Patients
N = 73
Prostate cancer
Breast cancer
Other cancers
3%
6%
9%
3%
4%
1%
3%
3%
6%
6%
Figure 3. Proportion of patients experiencing 1 or more SREs
during study.
DENOSUMAB FOR PROSTATE CANCER BONE METASTASES
513