GADOLINIUM CONTAINING CONTRAST AGENT PROMOTES
MULTIPLE MYELOMA CELL GROWTH: IMPLICATION FOR CLINICAL
USE OF MRI IN MYELOMA.
M. Fulciniti1,2,3, S. Swaminathan4, P. Nanjappa1, S. Amin1, P. Chevireddy4 , N. Gokden4, K.M. Hiatt4, W. High4, M. Shammas1,2, R. Prabhala1,2, T. Hideshima1, P. Tassone3, K. C. Anderson1, S.V. Shah4 and N. C. Munshi1,2
1Jerome Lipper Multiple Myeloma Center, Dana-Farber Cancer Institute, 2VA Boston Healthcare System, Harvard Medical School, Boston, MA 02115; 3University of "Magna Græcia", Campus Salvatore Venuta, Catanzaro 88100, Italy; 4 University of Arkansas, USA.
ABSTRACT
RESULTS
Bone marrow infiltration by myeloma cells and osteolytic bone lesions are the
major features of Multiple Myeloma. Magnetic Resonance Imaging (MRI) has been
In vitro growth promoting effect of Omniscan on MM cells.
used in MM not only to image bone marrow (BM) and to identify lytic bone disease
Omniscan exposure induces MM cell growth
The effect of Omniscan on growth of MM cell lines and primary cel s
but to also evaluate therapeutic response and prognosis. Gadolinium (Gd)-based
Omniscan was not able to overcome cytotoxic effects of
in a dose and time dependent manner
is enhanced in the presence of BMSC
conventional and novel agents in MM
contrast agents are frequently used to enhance MRI resolution. We evaluated
effect of the most common Gd-containing agent, Omniscan, on myeloma cells. We
600000
MM1S
INA6
30000
6000
6000
INA6
observed that Omniscan induced both time and dose dependent MM cell growth in
20000
KMS 12BM
400000
4000
INA6
4000
80000
LR5
XG1
10000
PBS
l
2000
2000
200000
MM1S
vitro (8-20 fold increase relative to control). Importantly, the presence of BMSC
ro
160000
100000
OMNISCAN
800
800
800
m 120000
U266
cont
enhanced the effect of Omniscan on growth of both MM cel lines and primary
400
400
400
of
cp
80000
%
40000
200
200
200
40000
cpm
MM cells. However, Omniscan was not able to overcome cytotoxic effects of
40000
50000
100
100
100
0
0
0
20000
conventional and novel agents in MM. This growth promoting effects were not
0
0.1
1
10
100
0
0.1
1
10
100
0
01
1
10
100
0
0
10
100
0
10
100
0
10
100
0
10
100
0
10
100
0
10
100
0
10
100
0
10
100
0
0
observed on normal BM stromal cel s. Evaluating the molecular mechanism of
0
REV DEX VEL
0
REV DEX VEL
0
REV DEX VEL
0
REV DEX VEL
2500
2500
1 day
2500
BM SC
-
+
-
+
-
+
-
+
2000
2000
action of Omniscan on MM cells, we observed time dependent ERK1/2
XG1
2000
3 days
OMNISCAN Dose (M)
RPMI 8226
MM1S
1500
l
1500
1500
Most patients receive Velcade, Revlimid or Dexamethasone as
5 days
1000
tro
1000
phosphorylation as well as reversal of growth promoting effects of Omniscan by
n
1000
o
500
500
Primary CD138+ myeloma cells
To analyze the effect of Omniscan in the context of
therapeutic agents for treatment of myeloma. To evaluate if
c
500
f
6000
specific inhibition of ERK signaling; however, Omniscan had no effect on STAT3
o
400
400
the bone marrow microenvironment, MM cells were
gadolinium contrast would antagonize the beneficial effects of
400
%
200
200
these agents in myeloma, we performed in vitro drug
200
incubated in the presence of BMSC at various
4000
and AKT signaling pathways. Next, we investigated in vivo effect of Omniscan in a
M
0
0
0
concentrations of Omniscan for 72 hours at 37°C. The
sensitivity assay. INA-6 and MM1S cells were cultured in the
0
0.1
1
10
100
0
0.1
1
10
100
CP
murine xenograft model of MM. Following detection of tumor, mice were treated
0
0.1
1
10
100
2000
presence of BMSC enhanced the effect of Omniscan
presence of Velcade (5nM), Revlimid (5 uM) or dexamethasone
with either iv Omniscan or PBS. Treatment with Omniscan significantly induced
1000
on MM cel growth even at lower doses (0.01 and 0.1
(1uM) with or without Omniscan (0.1 mM) for 72 hours, and cel
800
250
0
OPM1
0
10
100
0
10
100
800
l
BMSC
-
+
mM) in both MM cel lines (above) and primary MM
proliferation was measured by [3H] Thymidine uptake.
MM tumor growth compared to control mice (1042 ±243 mm3 vs 502 ±137 mm3
200
U266
ro
OPM2
l 600
600
OMNISCAN dose (M)
nt
ront
150
cells (left). This growth promoting effects were not
respectively; p=0.0001). Finally in autopsies in 8 MM patients with repeated
co
400
400
of
co
observed on normal BM stromal cells.
%
of
100
%
exposure to Omniscan, we quantified gadolinium in various tissues using
200
200
50
0
Inductively-coupled mass spectrometry. We observed massive quantities of
0
0.1
1
10
100
0
0
0
0.1
1
10
100
0
0.1
1
10
100
Omniscan activates ERK pathway in MM cells.
gadolinium accumulation in tissues of these MM patients regardless of their renal
A complex signaling network consisting of the IL-6R/Stat3, Ras/MAPK,
PI3K/Akt, notch, Wnt and Nf-kB pathways sustains MM cel growth and
function. These results, confirming both in vitro and in vivo growth promoting
Myeloma cell proliferation was evaluated by [3H] Thymidine uptake
MM1S
LR5
MM1S
800
PBS
survival. To evaluate the molecular mechanism involved in the induction
effects of Gd-containing contrast agent on MM, suggest the need for further
upon exposure to various concentrations of Omniscan at various time
OMNISCAN
points Results show induction of cell proliferation by Omniscan in a
Time (hrs)
0
24
48
72
0
24
48
72
of MM cell proliferation by Omniscan, MM1S and LR5 MM cells were
600
l
analysis of the mechanism of its action on myeloma cells and careful analysis of
ro
dose and time-dependent way in al the myeloma cel lines tested.
p-ERK1/2
exposed to 100 uM of Omniscan and whole cell lysates subjected to
400
its clinical impact in MM patients undergoing MRI evaluation.
fcont
immunoblotting. We observed time-dependent ERK1/2 phosphorylation.
ERK1/2
o
%
200
However, Omniscan had no effect on STAT3 and AKT signaling
p-Akt
pathways. Specific inhibition of ERK signaling pathway with U0126 was
p-STAT3
0
Cnt
5
20
Cnt
5
20
able to reverse growth promoting effects of Omniscan.
MATERIALS & METHODS
U0126 Dose (uM)
Cell proliferation was evaluated by Thymidine uptake.
Gadolinium accumulation in MM patients.
Western blot analysis was performed using Abs against Akt, STAT3, ERK, p-
In vivo growth promoting effect of Omniscan on murine model of MM.
Akt, p-STAT3, p-ERK.
Patient
Liver
Spleen
Heart
Bone
Lung
Kidney
Brain
marrow
All the experiments were performed at least in triplicate and results are
5000
)
1
81.41
39.28
12.66
51.09
1.12
78.89
N/A
expressed as mean +/- SD.
3 m 4000
m
2
6.48
N/A
2.67
3.37
1.08
9.51
0.44
(e
CNT (n=4)
To quantify gadolinium, quantitative measurement via inductively coupled plasma
3000
Next, we investigated in vivo effect of Omniscan in a
p=0.015
m
TR (n=4)
lu
3
9.85
3.73
2.77
0.41
1.96
4.42
0.39
o
mass spectrometry (ICP-MS) was performed.
murine xenograft model of MM in which LR5 cells were
2000
V
4
0.97
2.26
1.44
3.07
0.30
10.53
0.19
or
injected subcutaneously in SCID mice. Following detection
1000
LR5 cells-bearing mice received either a single daily 100 uL s.c. injection
Tum
5
11.80
23.48
5.57
N/A
2.65
13.00
1.35
of tumor, mice were treated with either i.v. Omniscan or
containing 100uM of Omniscan or 100 uL of PBS(control) for 1 week or one
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
PBS either once every 3 days for 3 weeks (upper figure)
6
15.24
10.47
2.74
8.12
0.78
24.54
0.85
Days from tumor detection and treatment
injection every 3 days for 3 weeks. Tumors were measured in two perpendicular
or once per day for 1 week (lower figure). Treatment with
7
5.06
1.05
1.36
N/A
0.25
1.47
3.74
Omniscan significantly induced MM tumor growth
dimensions once every 3 days.
8
66
179
15
415
4
34
N/A
1500
compared to control mice (1042 ±243 mm3 vs 502 ±137
)3m
mm3 respectively after 1 week of treatment; p=0.0001).
8 patients with MM that were exposed to gadolinium contrast at least once. In invidual with normal renal function,
m
TR (n=4)
( 1000
ICP-MS on the excised tumor samples confirmed
REFERENCES
e
CNT (n=4)
only negligible amounts of gadolinium is retained in the body and tissue levels of gadolinium should not exceed 1
mlu
p=0.0001
o
significant accumulation of gadolinium (60 mcg/g of dry
µg/gram of dry tissue ( or parts per million (PPM). In contrast, we found highly significant amount of gadolinium in
V
500
or
tissue) in the tumor tissues.
various organ tissues of MM patients and this was regardless of race, subset of myeloma or their level of renal
S. Swaminathan and S. V. Shah. New Insights into Nephrogenic Systemic
Tum
function. The highest amount of gadolinium detected in liver was 81 µg/gram, spleen was 179 µg/gram, heart was 15
Fibrosis. J Am Soc Nephrol 18: 2636­2643, 2007
0
0
1
2
3
4
5
6
7
8
µg/gram, bone marrow was 415 µg/gram, lungs was 4 µg/gram, kidney was 79 µg/gram and brain 4 µg/gram. None of
S. Swaminathan et al, Nephrogenic Systemic Fibrosis, Gadolinium, and Iron
Days from tumor detection and treatment
these patients had any clinical or histological evidence of Nephrogenic Systemic Fibrosis (NSF) based on their
Mobilization. N engl j med, 2007
CONCLUSIONS
available clinical history or detailed histologic analysis of the autopsy tissues.
M.A. Sieber et al. A Preclinical Study to Investigate the Development of
Gadolinium-containing contrast induces both in vitro and in vivo MM cell growth via ERK pathway activation.
Nephrogenic Systemic Fibrosis: A Possible Role for Gadolinium-Based Contrast
Effect of Gadolinium-containing contrast is accentuated in the presence of BMSC
Media. Investigative Radiology. 2008
Massive quantities of gadolinium accumulation is observed in tissues of MM patients regardless of their renal function.
These results suggest the need for further analysis of the mechanism of its action on myeloma cells and careful analysis of its clinical impact on MM patients undergoing MRI evaluation.

---