Prognostic impact of post-
l
transp antation
l
comp ete
remission (CR) in multiple myeloma (MM). Final
results of a p
fprospecti
p
ve study in a series of
yf
homogenously treated patients.
J. Martinez
Martinez Lopez
-
, A. Sureda
Sureda, J. Blade
Blade, J. de la Rubia, E. Albizua, M V
. . Mateos, R.
Martinez, J.M. Ribera, J. Garcia-Larana, F. de Arriba, M.T. Hernandez, M.J. Terol,
D. Carrera, J. Besalduch, F. Casado, L. Palomera, L. Escoda, S. Gardella, P. Ribas, P.
Fernandez, P. Fernandez-Abellan, B. Hernandez, J. San Miguel & J.J. Lahuerta
On behalf of the Grupo EEspañol de Mieloma (GEM)
H. Universitario 12 de Octubre. Madrid. Spain
XI International Myeloma Workshop. Kos, Greece, June 27, 2007
Background
EBMT Myeloma Sub-committee redefined response criteria for
MM treatment in 1998. In summary, negative immunofixation
(IF) was established as a requissite for CR.
At that moment there was not clear evidence of the prognostic
impact of achieving CR by IF.
In 2000, a retrospective study of Spanish Registry of MM
Transplantation, concluded that patients with negative IF
constituted a subgroup with better prognosis than those with
negative electrophoresis but positive IF.
IFM group identified VGPR as a group of response with
prognostic significance
Aim
A prospective study to validate
The si i
gn fi
ificance f
o response
d
measure by IF ft
a er
transplantation
The prognostic significancee of near complete response
defined as electrophoretic CR
This purpose was one of the main objectives in Spanish clinical
trial Myeloma 2000.
Patients andd Methods:
Methods
Definition of response
Complete response (CR): whhen IF results were negative
Electrophoretic Complete Response (e/nCR): when EEF(-)
but IF(+)
Partial Response (PR):
Minor Response (MR):
di
accor ng to EBMT it
cr
i
er a
Stable Disease (SD):
Progression (P):
Response was evaluated after chemotherapy, 1st or 2nd ASCT
d
an
d
consi
d
ere at h
t e moment of
of maximum response
Patients:
1088 patients were included in Myeloma 2000 clinical trial
between the years 2000 and 2004
740 patients were assessed for response after transplantation
CR (If --)=
)= 295 (39.9%)
e/nCR (EF+ If
If--)=
)= 129 (17.4%)
%
PR= 239 (32.3%)
MR= 23 (3%)
These groups were not analysed
SD= 6 (0.8%)
P=
P 49 (6.6%)
Patients:
740 patients were assessed for response after transplantation
Characteristic of patients:
p
Sex (M/F) %
57/43
Age media (range)
57.4 (30.82-70.02)
M componet
-
type (G/A//L)%
54/25/19
B2 microglobuline mg/L
4.26 (±3.8)
Hemoglobin g/L
10.8 (±2.2)
Calcium g/dL
9.8 (±3.3)
Creatinine uM
130 (±126)
Albumine g/L
36 (±7.1)
ECOG >2
44%
ISS (I/II/III)%
31/41/19
Durie Salmon (I/II/II) %
8/38/55
Plasmocytosis %
43.5 (±27)
Patients and Methods:
Clinical trial
All patients included in this analysis were treated with:
6 alternative chemotherapy cycles (using VBMCP/VBAD)
ASCT either BUMEL or MEL200 as conditioning regimen.
A second transplantation was performed if patients did not reach
e/nCR.
1088 p
VBMCP/VBAD (4 cycles)
+
Not progression
PBSC Harvest
VBMCP/VBAD (2 cycles)
99% (723p)
BUMEL/MEL200+ASTC-1
RC (If --)) or (EF --))
Follow
Follow--up
up and
Maintenance
Maintenanc therapy
16,2% (123p)
No RC (EF +)
2d
2 ASCT
Patients and Methods:
Statistical analysis
Main end point was PFS
Survival was estimated by Kaplan-Meier; differences were
examinated by
my Lo
L g-rank
g
test
Cox Regression was used to evaluate multivariable
prognostic
pg
models
Logistic regression was employed to study variables
associated to achieve CR (IF-)
Results
d
Me ian duration f
o f l
ollow-up for survivors was 45 months (7-90)
Univariate PFS analysis by response categories
1,0
CR
n/eCR
0,9
PR
0,8
P
0,7
P, between 0.01 y <0.001
Survival
median, m (IC) PFS % (5y)
0,6
eeFr
CR
60 (51-69)
50
0,5
e/nCR 49 (39-58)
29
0,4
Progression
PR
36(32-40)
21
03
0,3
P 0.01
0,2
P10 (9-12)
0
0,1
0,0
0
365
730
1095
1460
1825
2190
2555
2920
Days from diagnostic
Results
d
Me ian duration f
o f l
ollow-up for survivors was 45 months (7-90)
Univariate PFS analysis CR vs e/nCR
1,0
CR
n/eCR
0,9
0,8
median, m (IC) PFS % (5y)
0,7
CR
60 (51-69)
50
Survival 0,6
reeF
e/nCR 49 (39 58)
-
29
0,5
0,4
Progression
P<0.001
0,3
,
0,2
0,1
0,0
0
365
730
1095
1460
1825
2190
2555
2920
days from diagnostic
Results
d
Me ian duration f
o f l
ollow-up for survivors was 45 months (7-90)
Univariate PFS analysis e/nCR vs PR
1,0
e/nCR
0,9
PR
0,8
0,7
median, m (IC) PFS % (5y)
0,6
eeSurvival
e/nCR 49 (39-58)
29
Fr 0,5
PR
36(32-40)
21
0,4
Progression 03
0,3
P 0.01
0,2
0,1
0,0
0
365
730
1095
1460
1825
2190
2555
2920
days from diagnostics
Results
Univariate OS analysis by response categories
1,0
CR
e/n CR
0,9
PR
0,8
P
0,7
median, m (IC) OS % (5y)
p=0.07
0,6
Survival
CR
NR
82
p<0.001
05
0,5
e/nCR NR
79
0,4
Overall
p<0.001
PR
66 (50-82)
36
0,3
0,2
P14 (9-12)
10
0,1
0,0
0
365
730
1095
109
1460
146
1825
182
2190
219
2555
255
2920
292
Days
Results
Multivariate analysis Cox regression model
Variables associated to PFS in pa
patients evaluated for resp
fponse
(model 2=147.867 p<0.0001)
Variable
SS
RR
95% IC
Response
p<0.0001
CR
p<0.0001
1
/
n e CR
p=0.006
1.558
(1 1
. 3-2.1)
PR
p<0.0001
2.19
(1.7-2.8)
P
pp<0.0001
0001
11 21
.
(6 7
. -
7 18.69)
Hb>11g/L
p=0.01
0.73
(0.58-0.9)
ECOG>2
p=0.001
p
1.56
(1.15-2.1)
Durie-Salmon
p=0.004
1,39
(1.14-1.69)
Results
Multivariate analysis Cox regression model
Variables associated to OS in pa
p tients evaluated for resp
fponse
(model 2=119.92 p<0.0001)
Variable
SS
RR
95% IC
Response
p<0.0001
CR
p<0.0001
1
/
n e CR
0.07
1.4
PR
p<0.0001
2.47
(1.7-3.4)
P
pp<0.0001
0001
10 49
.
(5 9
. -
9 20.3)
ISS
p=0.01
1.43
(1.16-1.74)
ECOG>2
p=0.001
p
1.98
(1.3-1.9)
Durie-Salmon
p=0.006
1.47
(1.11-1.8)
Results
Logistic regression analysis to identify variables associated to
obtain CR IF- (model 2=44.84 p<0.0001)
Variable
SS
RR
95% IC
M-component<30g/L
0.001
2.4
(1.5-3.8)
Conclusions
Patients with CR IF- have longer PFS than patients with e/nCR
(EEF-/IF )
+ , nd
a
th s
e e, l n
o ger th
than
an PR (EBMT c i
r t
it i
er )
a pati
ti n
e ts
nts.
This large
i
prospect ve study shows h
t at response after
transplantation is a mayor prognostic factor that could be used
as a surrogate marker to predic
gp
t disease outcome.
In the near future more accuratee methods to define response
are needed, as IF- includes an heterogeneous group of patients,
with either very good or bad prognosis.
XI International Myeloma Workshop. Kos, Greece, June 27, 2007