RANKL, MIP-1:
Clinical Implications in Myeloma
10th International Myeloma Workshop;
Sydney, Australia; 10-14 April 2005
vangelos erpos,
erpos, MD, PhD
Faculty of Medicine Imperial College, London, UK &
General Airforce Hospital, Athens, Greece
Introduction
· Most frequent
presentation of
the disease
(90%)
· Osteolytic
lesions (60%),
diffuse
osteopenia (10-
15%),
· very rarely
increased bone
formation
MM bone disease: pathogenesis
Skeletal destruction
results from
increased osteoclastic
activity, which is not
accompanied by a
comparable increase
in bone formation
LYTIC LESIONS
BONE MARROW
osteoblasts
OPG
Myeloma Cells
(-)
RANKL
CD138
(?)
IL-6
DKK1
41 integrin
RANKL
RANKL
VCAM-1
VCAM-1
BMSCs
RANKL
IL-11, IL-1,
(-)
bFGF
OPG
IL-6
TNF, M-CSF
RANK
MIP-1, IL-3, HGF
Osteoclast
Activated osteoclasts
precursor
TRACP-5b
Collagen type-1
degradation products:
Bone resorption
NTX, ICTP, CTX
Bone matrix
Osteoclast activation is important
for MM cell survival
After 14 days
Abe et al, Blood 2004;104:2484
RANK RANKL - OPG pathway (I)
· RANKL
RANKL (receptor
activator of nuclear
factor-
factor-B ligand):
·M
·
ember
Member of
of TNFs
TNFs
superfamily
· Induces
Induces
differentiation and
proliferation of
osteoclast
progenitors
·D
·
ecreases
Decreases
apoptosis of
mature osteoclasts
Hofbauer et al; Cancer 2001;
92: 460
OPG
OPG is encoded by
a single gene on
chromosome 8q24
It is mainly secreted
by stromal cells
OPG is the decoy
receptor for RANKL
that blocks the
RANKL-RANK
interaction and thus
inhibits osteoclast
differentiation and
function
Ashcroft et al , Lancet Oncol 2003;4:284
Serum OPG in MM patients
150
100
75
100
50
L
50
m
25
15
ng/mL
15
ng/mLng/
10
OPG
10
OPG
5
5
0
0
0
1
>1
Myeloma
Control
Control
Number of radiographic lesions
75
50
L
/m
25
g/m
15
nG
10
OP
5
0
Seidel et al,
0
1
2
3
4
Blood
WHO
WHO performan
performance status
2001;98:2269
RANKL in bone marrow biopsies
Roux et al , BJH 2002;117:86
RANKL
OPG
Normal
MM with osteolysis
Giuliani et al, Exp Hematol 2004;32:685
sRANKL/OPG ratio in newly diagnosed MM
2.0
30
p<0.0001
1.5
121
36
1.0
52
.5
0.0
L/OPG
-.5
68
134
sRANK
-1.0
159
149
158
Log
-1.5
N =
45
10
28
83
Control
A
B
C
Bone Disease
Terpos et al, Blood 2003;102:1064
sRANKL/OPG ratio and markers
of bone resorption in MM
2.8
1.8
r=0.87, p<0.0001
1.6
2.6
1.4
2.4
1.2
2.2
1.0
2.0
.8
1.8
.6
1.6
-5b .4
1.4
.2
r=0.69, p<0.0001
TRACP
1.2
0.0
LGNTX
LG
-1.5
-1.0
-.5
0.0
.5
1.0
1.5
2.0
-1.5
-1.0
-.5
0.0
.5
1.0
1.5
2.0
LGRATIO
LGRATIO
Terpos et al, Blood 2003;102:1064
sRANKL/OPG ratio and markers of
disease activity in MM
1.6
2.5
r=0.68, p<0.0001
1.4
2.0
1.2
1.5
1.0
.8
1.0
.6
.5
.4
r=0.52, p<0.0001
.2
LGIL6 0.0
LGBET2MG
-1.5
-1.0
-.5
0.0
.5
1.0
1.5
2.0
-1.5
-1.0
-.5
0.0
.5
1.0
1.5
2.0
LGRATIO
LGRATIO
Terpos et al, Blood 2003;102:1064
Prognostic value of
sRANKL/OPG ratio in MM (I)
sRANKL/OPG <1 (n=35)
100
%
89%
80
60
survivalof
sRANKL/OPG 1-3 (n=54)
40
32%
20
sRANKL/OPG >3 (n=32)
Probability
20
0
12
24
36
48
60
Months post diagnosis
Terpos et al, Blood 2003;102:1064
Prognostic value of
sRANKL/OPG ratio in MM (II)
Parameter
N
Relative risk
P-value
Prognostic
(95% CI)
score (points)
sRANKL/OPG
<1
35
1.00
1
1-3
54 7.8 (2.5-23.7) <0.001
2
>3
32 24.7 (7.2-84.4) <0.001
4
CRP10
89
1.00
2
>10
32
2.9 (1.6-5.2)
0.001
3
-microglobulin
57
1.00
0.043
1
2 3
64 2.2 (1.03-4.9)
3
>3
Prognostic value of
sRANKL/OPG ratio in MM (III)
HAMMERSMITH PROGNOSTIC INDEX
Prognostic score <6
<6 (n=26)
(n=26)
100
96%
%
80
P<0.0001
60
survival
Prognostic score 6-8
6-8 (n=55)
52%
of
40
20
Prognostic score >8
>8 (n=40)
(n=40)
Probability
20
0
12
24
36
48
60
Months post diagnosis
Terpos et al, Blood 2003;102:1064
HAMMERSMITH PROGNOSTIC INDEX
Prognostic
Prognostic scor
score <6 (n=26)
26)
100
96%
l%a
80
rvivu
P<0.0001
60
fs
Pro
Pr
Pr gnosti
ognosti
ognost c
ic sco
sco re
ore 6-
6- 8 (n=55)
(n=55)
52%
yo
40
ilit bab 20
rob
Prognostic score >8
>8 (n=40)
(n=40)
P
20
0
12
24
36
48
60
Months post diagnosis
Bataille's Prognostic
rognostic Sy
Sy stem
CRP and B2 <6 (n=
(n= 65)
100
%
80
73%
rvivalu 60
s
P<0.0001
CRP
CRP or B2
B2 >6
>6 (n=39
(n=3 )
9)
40
lityof
22%
CRP and B2 >6 (n=
(n= 17)
20
robabiP 20
0
12
24
36
48
60
Months post diagnosis
sRANKL/OPG ratio post ASCT
Alterations in median values of OPG and sRANKL
post-ASCT
post-ASCT
9
8
*p<0.03
7
*
6
p
m
o
l/L
5
OPG
pmol/L 4
sRANKL
pmol/L
3
*
2
1
0
1
2
3
4
5
6
7
8
9
10
11
12
13
Months post-ASCT
Terpos et al, Leukemia 2004;18:1420
sRANKL/OPG ratio post Thal-Dexa (I)
70
60
50
*
)
40
-2
x10( 30
atio
p<0.01
r
20
p<0.0001
compared to baseline
KL/OPG
10
AN
sR
0N=
30
35
35
35
Controls
Baseline
At 3 mont
ont hs
At 6 months
onths
Terpos et al, IMW Sydney 2005
sRANKL/OPG ratio post Thal-Dexa (II)
60
r=0.541, p=0.006
50
ratioG 40
KL/OP
30
KL/OP
20
sRANof 10
(%)
0
-10
change
-20
-3
-2
-1
0
1
2
3
4
Change (%) of CTX
Terpos et al, IMW Sydney 2005
sRANKL/OPG ratio post Thal-Dexa (III)
60
r=0.581, p=0.001
50
ratioG 40
KL/OP
30
20
sRANof 10
(%)
0
change
-10
change
-20
-2
-1
0
1
2
3
4
5
change (%) of TRACP-5b
Terpos et al, IMW Sydney 2005
sRANKL/OPG ratio post Thal-Dexa (IV)
100
p=0.112
% 80
60
survivalof
sRANKL/OPG <15x10-2 (n=26)
40
20
Probability
sRANKL/OPG >15x10-2(n=9)
0 0 6 12 18 24 30 36 42 48 54 60
Months post treatment
Is RANKL produced by MM cells?
Giuliani et al, Haematologica 2004;89:1118
Lai et al, BJH
2004;126:192
**
1.40
1.20
/L)
pM( 1.00
0.80
nlevels
0.60
mea
0.40
RANKL
0.20
0.00
BMSC
BMSC+XG-1
XG-1
Giuliani et al, Haematologica 2004;89:1118
180
300
*
*
160
250
140
control
control
vs
120
vs
200
100
*
*
150
80
variation
60
100
-6variation
-11
IL
40
IL
IL
50
of
20
of
%
%
0
0
BMSC
BMC-1
BMSC
BMC-1
RANKL
RANKL + RANK-Fc
2.5
2.0
1.5
1.0
LGIL6
.5
Giuliani et al, Haematologica
0.0
2004;89:1118
-1.5 -1.0
-.5
0.0
.5
1.0
1.5
2.0
LGRATIO
MIP-1 and bone disease in MM
· MIP-1 is a member of the CC chemokine
family and is primarily associated with cell
adhesion and migration
· i
·is chemotactic for monocytes and
and
monocyte-like cells, including osteoclast
precursors
· i
·is produced by myeloma cells and
and directly
stimulates osteoclast formation and
and
differentiation in a dose dependent way
Uneda et al, BJH 2003;120:53
160
120
*
*
MNC/Well
80
*
250
Positive
*
40
200
23c6
*
0
150
MNC/Culture
*
D3
0
5
50
200
-1 (pg/mL)
100
Positive
23c6
50
360
300
*
0
MNC/Well 240
*
0
5
10
50
180
RANKL (ng/mL)
120
Positive
600
23c6
Media 10 100 10 100 0 IL-6 (pg/mL)
0
0 200 200 200 MIP-1 (pg/mL)
Han et al, Blood 2001;97:3349
20
15
*
(mg/ml)
10
lgG2b
5
Serum
0
ctrl lgG + PBS
anti-MIP-1
Oyajobi et al, Blood 2003;102:311
Serum MIP-1 in MM patients (I)
3
p<0.01
85
86
2
1
-1 0
5
3
4
MIP
2
Log -1N=
10
27
48
Group A
Group B
Group C
Bone Disease
Terpos et al, BJH 2003;123:106
Serum MIP-1 correlates with bone resorption
2.8
r=0.594, p<0.0001
2.6
2.4
2.2
2.0
1.8
1.6
1.4
NTX 1.2-1.0 -.5 0.0
.5
1.0
1.5
2.0
2.5
3.0
MIP-1
Terpos et al, BJH 2003;123:106
1.8
r=0.376, p<0.0001
1.6
1.4
1.2
1.0
.8
.6
5b
.4
-
.2
TRACP 0.0-1.0 -.5 0.0 .5 1.0 1.5 2.0 2.5 3.0
MIP-1
Terpos et al, BJH 2003;123:106
Serum MIP-1 correlates with sRANKL levels
2.0
r=0.382, p<0.0001
1.5
1.0
.5
0.0
sRANKL
-.5-1.0 -.5
0.0
.5
1.0
1.5
2.0
2.5
3.0
MIP-1
Terpos et al, BJH 2003;123:106
Serum MIP-1 correlates with 2M
1.0
r=0.401, p<0.0001
.8
.6
G
.4
2-
.2-1.0 -.5
0.0
.5
1.0
1.5
2.0
2.5
3.0
MIP-1
Terpos et al, BJH 2003;123:106
Predictive value of serum MIP-1
100
%
80
MIP-1 48 pg/mL (n=76)
60
survivalof
40
P=0.021
MIP-1 > 48 pg/mL (n=9)
20
Probability
0 0 1 2 3 4 5 6 7 8 9 10
Years post diagnosis
Terpos et al, BJH 2003;123:106
CONCLUSION
· Myeloma
Myeloma bone disease is due to increased
osteoclast activity and abnormal osteoblast
function.
· RANK/RANKL/OPG
RANK/RANKL/OPG and MIP-1 pathways are
important not only for the pathogenesis of
myeloma bone disease but also for myeloma
cell
survival,
being
targets
for
the
development of novel anti-myeloma agents
Aknowledgments
Hammersmith (London)
Greece:
Amin Rahemtulla
M. A. Dimopoulos (Un. Athens)
Marianna Politou
A. Anagnostopoulos (Un. Athens)
Richard Szydlo
K. Zervas (Thessaloniki)
Chrissy Giles
D. Mihou (Thessaloniki)
Diana Samson
K. Tsimirika (Thessaloniki)
John Goldman
K. Tsionos (Athens)
Jane Apperley
E. Voskaridou (Athens)