Evidence for Cytogenetic and Fluorescence in-situ Hybridization (FISH) Risk-Stratification of Newly Diagnosed Multiple Myeloma in the Era of Novel Therapies*
Prashant Kapoor,
p, MD, Rafael Fonseca, MD, S. Vincent Rajkumar
j
, MD, Shirshendu Sinha, MBBS, Morie A. Gertz, MD, A. Keith Stewart, MD, P. Leif Bergsa
,g gel
g , MD, Martha Q.
Q Lacy,
y, MD, David Dingli
g , MD, PhD, Rhett P. Ketterling,
g, MD, Robert A. Kyle
y , MD,
Thomas E. Witzig, MD, Philip R. Greipp, MD, Angela Dispenzieri, MD, Shaji Kumar, MD.
Division of Hematology , Mayo Clinic, Rochester, MN
BACKGROUND
RESULTS
FIGURES
DISCUSSION AND CONCLUSION
Heterogeneity in multiple myeloma (MM) leads to
Two hundred and ninety MM patients had CG and PCLI performed within 90 days
The mSMART model integrates risk assessment by three tests,
disparate outcomes with a wide variation in overall
prior and up to 7 days of initiation of anti-myeloma therapy. In addition, interphase
1A
1B
two of which are increasingly being performed in the community as
surv
su iv
rv al
ival (OS).
FISH results were available on all patients.
2A
2B
ii
in ti
iti l
a work-up of
l
new y di
diagnos d
e MM
MM (N
( D
N MM)
D
.
P=0.003
Based upon the evidence predominantly gathered from
P=0.002
P=0.2
P=0.01
The median age at presentation was 64 years and 177 (61%) were males. Two
Our study validates high-risk features defined by FISH and CG in
newly diagnosed patients treated with standard
Std.- Risk by CG
hundred and thirty-six (81%) patients had received novel agents, primarily
Std-Risk
Std-risk
Std-risk by CG
the risk stratification model among patients treated since the
chemotherapy and/or stem cell transplantation (SCT),
High-Risk
thalidomide (n=50), lenalidomide (n=199) and bortezomib (n=79) as front-line or
introduction of novel therapies. Either test identifies a high-risk
the current Mayo (mSMART) prognostic model uses a
High-risk
High-risk by CG
salvage therapies.
High-r isk by CG
cohort with an HR of approximately 2.
combination of metaphase cytogenetics (CG),
Median OS: NR vs.37 months
Median OS: NR vs.31 months
median OS 35 vs.16 mont hs
Median OS : NR vs. 37 months
fluorescence in-situ hybridization (FISH) and plasma cell
One hundred and eighty-five patients (64%) had received a novel agent-based
Particularly, in patients stratified as standard-risk by one of the two
labeling index
index (PCLI
(PCL )
I) for
fo risk
risk-strati
stra f
ti icat
ca iton
o of
of MM
MM.
front-line therapy.
tes
te ts,
sts, inform
o
ation
ation provided
provided by the
th ot
other appears to
to be
be of
of additional
Time in Month s
1C
1D
2D
prognostic value.
The prognostic value of this model has not been tested
One hundred and twelve patients (39%) had undergone autologous SCT, of which
P=0.001
2C
P=0.3
P=0.3
P=0.01
in the era of expanded use of novel agents.
thirteen received a second SCT.
Std- risk by FISH
Our study has reinforced the importance of utilizing CG and FISH,
Std-risk
Std-risk
both of which appear to be independent prognosticators at
The objective of our study was to evaluate the
The median estimated OS for the cohort was 65 months and the median
Std-risk by FISH
High-risk
diagnosis and complement each other for risk-analysis.
significance of this stratification model in the era of novel
estimated follow-up was 29 months from initiation of therapy.
High-risk
High- risk by FISH
High-risk by FISH
therapies, and to assess the independent prognostic
Median OS: 65 vs. 29 months
Median OS: 65 vs. 37 months
Median OS: 37 vs. 16 months
OS: NR vs. 35 m onths
Surprisingly, the slide-based PCLI that has hitherto been an
Eighty (28%) patients were deemed high-risk by any one or more of the three
value of each of the components in the model
extremely powerful independent prognosticator in different settings
tests:
te
CG
CG, FI
FISH
SH or
or PCLI
PCLI.
(l
(newly di
diagnos d
e ,
l
re aps d
e , pre- tl
transpl t
an , ch
i
ron c st b
a l
ble l
p t
a eau
Among these patients, 22 (8%), 51(18%) and 27 (9%) patients were considered
Figure 2A: Cytogenetics-based risk-stratification of patients who have high-risk
phase) was found to be a relatively ineffective prognosticator in
METHODS
high-risk by CG, FISH and PCLI, respectively.
features on FISH: Overall survival of FISH-detected high-risk patients (n=51) further
Figure 1: Overall survival (OS) of high-risk vs. standard-risk patients with newly
NDMM patients receiving novel agents.
subdivided into two groups based on presence (n=6) or absence (n=45) of high-risk features
Data were abstracted from a cohort of 1556 patients who
diagnosed multiple myeloma classified (A) by mSMART criteria: Median OS of high-
on cytogenetics: median survival 16 vs. 35 months, respectively 2B: Cytogenetics-based
Survival analyses are summarized in the table (below) and figures 1, 2 and 3.
risk patients (n=80) is 37 months vs. not reached (NR) for std-risk patients (n=210). (B) by
risk-stratification of patients who have standard-risk features on FISH: Std-risk patients
We advocate the use of both CG and FISH for risk-assessment of
presented between January 1999 and March 2009
FISH ( fluorescence in-situ hybridization) alone: Median OS for high-risk (n=51) vs.
by FISH (n=239) further subdivided into two groups based on presence (n=16) or absence
NDMM until gene expression profiling-defined high-risk signatures
standard-risk (n=239) patients is 31 months and not reached, respectively. (C) by
High-risk MM was defined by the Mayo stratification
(n=223) of high-risk features on cytogenetics: median survival 37 vs. not reached (NR),
metaphase cytogenetics alone: Median OS high-risk (n=22) vs. standard-risk (n=268) is
become more widely available for clinical practice. However, given
respectively. 2C: FISH-based risk-stratification of patients who have high-risk features
model as the presence of any one or more of the following:
29 vs. 65 months, respectively. (1 D) by PCLI alone: median survival of high-risk (n=27) vs.
on cytogenetics: High risk patients by cytogenetics (n=22) further subdivided into two groups
the fact that FISH allows identification of a larger proportion of high-
std-risk (n=263)
() is 65 vs. 37 months, respectively.
y
based on presence
pr
(n=6)
(n=6) or absence (n=16)
(n=16) of high r
- irisk featur
eat es
ur
on FISH
FI
:
SH: median survi
survival
v 16
CG: hypodiploidy, monosomy of chromosome 13 or
Hi h
g
ik
-ri
-r s
i k
s
Md
Me
M
i
Md
dian Sur
S
i
v val
Ris
Ri k rati
ttio
P Vl
-V
- a
V lue
l
risk
t
pa iti t
en s compar d
e
i
w thC
ith G
CG, th
the former sh l
ou db
ld be i
g ven
i
pr i
or tity
3A
vs. 37 months, respectively. 2D: FISH-based risk-stratification of patients who have
deletion 13q; FISH: deletion of
3B
p53 (locus 17p13) or
Stra
r ti
t fi
f c
i at
a io
i n by
b
(mo
(m nt
n hs
h )
standard-risk features on cytogenetics Std-risk patients by cytogenetics (n=268) further
if only one of the tests can be performed for any reason.
immunoglobulin heavy chain (IgH) translocations, t(4;14)
subdivided into two groups based on presence (n=45) or absence (n=223) of high-risk features
P= 0 . 0 0
Cyto
Cy g
to e
g neti
e c
ti s
c
29
2.2
2.
.02
.0
P=
02
0 . 0 0 0 2
P = 0. 0005
on FISH: median survival 35 months vs. not reached, respectively.
(p16.3;q32) or t(14;16)(q32;q23); PCLI 3%. Patients were
[Co
[C n
o tr
t o
r l (std
(s -r
td is
i k
s ):
)
CONFLICTS OF INTEREST
considered to be standard-risk if they lacked all the
65 mo
m nt
n h
t s]
· R Fonseca is a consultant for Genzyme, Celgene, BMS, Otsuka, Halozyme and Medtronic. His
aforementioned abnormalities.
Figure 3: Risk-stratification based on presence of high-risk features on
research is funded by Cylene and Proteolix.
M e d ia n O S : N R vs . 35 vs . 16 m o n t h s
M ed i a n O S : N R v s . 3 7 v s . 3 5 v s . 1 6 m o nt h s
Fluo
Fl
re
uo s
re ce
c nc
n e
c in-sit
s u
31
2.0
2.
.02
.0
FISH/cytogenetics (CG) or both. 3A Median OS is 16 months vs. 35 months
vs. not reached for patients with high-risk features on both (n=6), either (n=61)
· PL Bergsagel is on the advisory board of Amgen, Genentech and Celgene.
Survival estimates were created using the Kaplan-Meier
hyb
y ridizat
a ion
(Co
(C n
o tr
t o
r l:
l NR
N )
R
or neither (st
( d-risk;n=223),
;), respectively.
y P=0.0002. No difference in survival
T i me i T
n ime
mo n i
t n
h s mo nt h s
T i m e i n m o nths
· PR
PR Greipp and A Di
Di
i
spenz i
er re i
ce ve l
c ii
lini l
ca t il
rial f
d
un i
ding from Cl
Celgene.
method and compared by log-rank tests. Prognostic value
(FI
(F SH
S )
H
was noted between high-risk by either vs. high risk by both, likely a reflection
St a nd a rd -ri sk
S t a nd a rd -ri sk b y F I S H an d C G
of small number of the patients in the group with high-risk features on both
· MA Gertz received honoraria from Millennium and Celgene.
of the three tests was separately assessed in a
Pl
P a
l sma
s
cel
e l labe
b lin
li g
n
37
1.1
1.
.85
.8
H i gh-r is k by C G b ut s t d-ris k by F I S H
FISH and CG. 3B: Median survival is 16 months vs. 35 months vs. 37 months
H i gh-r is k by eit h e r FI S H o r C G
· S. Kumar receives research funding from Celgene, Bayer, Genzyme, Millennium and Novartis.
multivariable analysis using Cox proportional hazards
H i gh-r is k by F I S H bu t s t d -ri sk b y CG
vs. not reached for high-risk patients by both FISH and CG (n=6) , high-risk by
inde
d x (P
( C
P LI
L )
(Co
C n
o tr
t ol
r :
ol 65
6 mo
m nt
n hs)
hs
model. Additional prognostic contribution of FISH in
H i gh-r is k by bot h F I S H an d CG
CG only (n= 16), high-risk by FISH only (n=45) and standard-risk patients
H i gh -ri s k by F I S H an d CG
patients classified as high and standard-risk by CG, and
(n=223), respectively. Note that the survival curves of high-risk patients by
either FISH or CG are intertwined.
vice versa was evaluated.
*Results in this poster have been updated since ASH abstract submission in August, 2009

---