A New Staging System for Multiple Myeloma Based on the

From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
A New Staging System for Multiple Myeloma Based on the Number
of S-Phase Plasma Cells
By J.F. San Miguel, R. Garcia-Sanz, M. Gonzalez, M.J. Moro, J.M. Hernandez, F. Ortega, D. Borrego,
M. Carnero, F. Casanova, R. Jimenez, J.A. Portero, and A. Orflo
In the present study, we analyzed the cell cycle distribution
of bone marrow (BM) cells in 120 untreated multiple myeloma patientsusing a DNA/CD38 double-stainingtechnique at flow cytometryin which plasma cells (PCs) can be
clearly discriminated from residual BM cells based on their
CD38 expression. This approach allows us t o determine the
proliferative activity of both PCs and residual normal BM
cells. The percentage of S-phase cells in the myelomatous
population was found t o be significantly lower than that
of the residual normal BM cells ( P < .001). Regarding the
proliferative activity of myelomatous cells, patients with a
high number of S-phase PCs (>3%) showed a significantly
(P< .05) increased incidence of anemia and hypercalcemia;
higher values of P,-microglobulin (&M), urea, and creatinine; and higher numbers of peripheral blood natural killer
cells, as well as a poor prognosis as assessed both by response duration and overall survival. With respect t o the
residual BM normalfraction, a low proliferative activity was
significantly ( P < .05) associated with the presence of anemia and neutropenia together with increased numbers of
BM PCs, a higher incidence of Ben- Jones myelomas, and
DNA diploidy. Multivariate analysis showed that the
number
of S-phase PCs was the most importantindependent prognostic factor, allowing us t o discriminate two subgroups of
patients withdifferent prognoses, even within the same clinical stage. Moreover, the S-phase PCs, together with &M,
age, and performance status, represent the best combination of disease characteristics for stratifying patients according t o prognosis and allow theestablishment of a simpleandpowerfulstaging
system formultiple myeloma
patients. In addition, this classification can be used for planning treatment in patients who are candidates for transplantation.
0 1995 by The American Societyof Hematology.
I
NCREASED USE of intensive therapeutic approaches for
in part, this is probably related to technical artifacts as was
recently been pointed out by the DNA Consensus Conferthe treatment of patients with multiple myeloma (MM)
ence for the standardization of studies oncell DNA conmakes the identification of different prognostic subgroups
tent.""' Accordingly, it has been indicated that the influence
very important. This distinction would allow better patient
of several artifacts in the flow cytometric assessment of Scounseling and an improved evaluation of clinical trials.
phase cells in bone marrow (BM) of MM patients should be
Among the different prognostic factors identifiedin MM
resolved in advance. Of the factors that may influence the
patients, one of the most important is the proliferative activresults, the most relevant fact is that, in MM BM samples,
ity of plasma cells (PCs) -the proportion of PCs that synPCs coexist with residual normal cells that are usually prolifthesize
This has generally been assessed by the
erating. For this reason, some
have
used
simultalabeling index (LI) obtained after incubation of cells in the
presence of tritiated thymidine"' or brom~deoxiuridine."~ neous staining of DNA and cytoplasmic Ig to discriminate
PCs. However, although being of help for the identification
These techniques are relatively complex and time-consumof aneuploid DNA peaks, the separation between PCs and
ing and, in some cases, include the use of radioactive isonormal residual cells that this technique provides is not suftopes. Because of this, they have not been extensively used
ficient for cell-cycle analysis purposes.
and are still restricted to research laboratories, despite the
Immunophenotypical studies by
flow
cytometry have
proven clinical value of the L1 for both the identification
shown that PCs display a singularly high reactivity for the
of patient subgroups with different
and for the
CD38 antigen, which allows a clear distinction between PCs
differential diagnosis between MM and monoclonal
and the other BM cell population^.""^ Taking advantage of
gammopathies of undetermined ~ignificance.~.'
this strong reactivity for the CD38 antigen detected in PCs,
In recent years, flow cytometry has provided new insights
we have used a double DNAKD38 staining technique in
into the biology of neoplastic cells, including cell-cycle analwhich PCs can be clearly discriminated from residual BM
ysis. Although several studies have analyzed the cell-cycle
cells based on their CD38 expression. Thus, the cell-cycle
distribution in different hematologic malignancies, quite
distribution ofMM PCs can be calculated separately from
variable and discrepant results have been obtained; at least
that of the residual cells.I8 In thepresent study, we have
evaluated the proliferative activity of both myelomatous PCS
and residual normal BM cells in 120 untreatedmyeloma
From The Castelano-Leone5 (Spain) Cooperative Group for the
Study of Monoclonal Gammopathies; the Department of Hematolpatients, using an easy-to-perform, standardized and reproogy, University Hospital of Salamanca, University of Salamanca,
ducible method'x," for measuring the S-phase of both cell
Salamanca, Spain.
populations. The most important independent prognostic facSubmitted June 3, 1994: accepted September 13, 1994.
tor was the proportion of S-phase PCs, which together with
Address reprint requests to Jesus F. San Miguel, MD, PhD, SerB,-microglobulin@,M),
age, and performance status alvicio de Hematologia, Hospital Universitario de Salamanca, Paseo
lowed
the
establishment
of
a simple and powerful staging
de San Vicente, 58-182, 37007, Salamanca, Spain.
system for MM patients.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
"advertisement" in accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1995 by The American Society of Hematology.
0006-4971/95/8502-0134$3.00/0
440
MATERIAL AND METHODS
Patients. A total of 120 untreated symptomatic myeloma patients diagnosed according to the criteria of the Chronic-LeukemiaMyeloma-Task-ForceZowere included in the present study. Patients
Blood, Vol 85, No 2 (January 15), 1995: pp 448-455
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
PROGNOSIS IN MULTIPLE MYELOMA
were treated according to the protocol of the Spanish cooperative
group PETHEMA” which includes either melphalan and prednisone
(43% of cases) or alternating cycles of VCMPNBAP (57% of cases).
No significant differences were observed regarding the treatment
used.’’ The median survival for the whole series was 37 months.
In each patient, the most relevant clinical and laboratory disease
characteristics documented at diagnosis were evaluated for their
prognostic significance. These included clinical features (age, sex,
performance status according to the Eastern Cooperative Oncology
Group (ECOG) scale, bone pain, loss of weight, the presence of
hepato-splenomegaly and the existence of plasmacytomas), hematologic parameters (hemoglobin, white blood cell [WBC] and platelet
counts, and erythrocyte sedimentation rate [ESR]), serum biochemical data (creatinine, urea, calcium, lactic dehydrogenase [LDH], and
PZMlevels), electrophoretic characteristics (total proteins, albumin,
type of monoclonal (M) component and the presence of urine Ig
light chains), percentage of BM plasma cells (PCs), and the presence
or absence of bone lesions. In addition, patients were grouped into
clinical stages according to Dune and Salmon’s’’ criteria.
The median age of the cohort studied was 69 years (range, 38 to
91). with 62 males and 58 females. According to their clinical stages,
patients were distributed as follows: LA, 6%; IIA, 24%; IIB, 2%;
IIIA, 52%; IIIB, 16%. A serum creatinine 2 2 mg/dL was present
in 18% of patients. Serum calcium was greater than 11 mg/dL in
25% of patients.
A monoclonal serum component was found in 86% of patients:
IgC, 55%; IgA, 30%; IgD, 1%. The remaining 13% had Bence Jones
myeloma. In1% of the patients, no monoclonal serum protein or
monoclonal urine protein was identified. The serum monoclonal light
chain was K in 63% and X in 37%. Urine immunoelectrophoresis
was performed in all patients. In 45% of the cases, a urine monoclonal light chain was detected.
Criteria fur respunse. An objective response (OR) was defined
as ( l ) a reduction of 50% or more of the M component, (2) improvement in performance status by at least two grades, and (3) a decrease
of greater than 50% in the measured cross-sectional area of plasmacytomas. Furthermore, the size and number of lytic bone lesions
should not have increased, and correction of hypercalcemia (< 11.O
mgldL), anemia (>9 mg/dL),and hypoalbuminemia (>3.5g/dL)
should also have occurred. Those patients who fulfilled all the above
criteria, but who had less than 50% reduction of M component were
considered to have had a partial response (PR). When the criteria
for an OR or PR were not fulfilled, the case was considered a
treatment failure.” Patients who died before completion of the proposed treatment were considered early deaths. Within the group of
patients with objective response, those who had no detectable M
component and less than 5% of BM PCs were considered complete
responses. Finally, relapse was defined as an increase of greater than
50% from the lowest level of serum M component achieved with
the initial therapy, an increase in size or number of lytic bone lesions,
and the development of extraosseus plasmacytomas, anemia, and/or
hypercalcemia. Overall survival (OS) was considered from the moment of diagnosis to the moment of death, and response duration
(D),
from the moment at which response was obtained until the
relapse.
~ z levels.
M
Serum PzM levels were measured by a radioimmunoassay method (Immunotech SA, Marseille, France). The upper
limit in 20 sex- and age-matched healthy individuals was 1.6 pg/
mL. In the present series of MM patients, the mean value was 5.8
2 5.1 pglmL (median value, 4.1 pg/mL).
DNAploidy studies. To analyze the presence or absence of DNA
aneuploidy, the technique of Vindelov et al’’ was used, Measurements were performed on a FACScan flow cytometer (Becton Dickinson, San JosC, CA) for at least 10,OOO eventdsample, using the
449
CellFit software program (Becton Dickinson). In the present series,
56% of patients had DNA hyperdiploidy, and 2% were hypodiploid.
Cell cycle analysis. For cell-cycle analysis a double-staining
procedure using CD38 and propidium iodide (PI) was used. Briefly,
between 100 and 200 p L of diluted whole BM containing 0.5 to 1
X lo6 WBCL was placed in the bottom of the tube. Cells were then
incubated for 15 minutes at room temperature with 10 pL of the
GR7A4 (CD38) monoclonal antibody (MoAb), washed once (5 minutes, 1,900 rpm) in phosphate-buffered saline (PBS) and incubated
for another 15 minutes in the dark (room temperature) with an antimouse Igs MoAb (F[ab’]’ fragments) (Dakopatts, Copenhagen, Denmark). Afterwards, 2 mL of ammonium chloride was added and
cells incubated in the dark for another 10 minutes (room temperature). Cells were then washed once in 1 mL of sodium citrate buffer
and resuspended in 200 pL of sodium citrate buffer. Afterwards, 1.5
mL of solution B, containing RNAse was added and cells incubated
for another 10 minutes period in the dark atroom temperature.
Finally, 1.5 mL of solution C, containing PI, was added and another
incubation for at least 10 minutes in the dark (room temperature)
was performed. Measurements were performed within 30 minutes
after the technique was finished ona FACScan flow cytometer using
the CellFit software, and counting at least 10,ooO cellslsample. The
percentage ofboth CD38 strong-positive cells andthe remaining
cell populations were calculated after excluding cell doublets on
a FL2-AreaFL2-Width dot plot using the PAINT-A-GATE-PLUS
software (Becton Dickinson). The proportion of cells in the different
cell-cycle phases was calculated using the RFIT mathematical model
included in the CellFit software after excluding cell doublets and
gating on boththe CD38 strong-positive and remaining cells, as
previously reported.” To identify the cell cycle of PCs, we selected
the gate for CD38 strongly positive cells, where there are only PCs,
using the same software. As internal controls and in order to confirm
the type of cells included in these gates, we have performed sorting
experiments simultaneously on both CD38 strong-positive cells and
the remaining cells in four patients using a FACStar flow cytometer.
The sorting was based on the cell DNA content and reactivity for
CD38. Ninety-seven percent ? 2%of the cells on theCD38”’
sorted fraction were plasma cells, as assessed by both morphologic
and immunologic criteria (cIgk+ or cIgl’ cells); by contrast, the
percentage of plasma cells in the other cell sorted fraction was lower
than 1% in all four cases analyzed (0.6% 2 0.4%).” Moreover, in
aneuploid cases, only the strongly CD38-positive cells showed an
abnormal DNA content. An additional theoretical problem with this
technique could be the existence of MM cases lacking a strong CD38
positivity; however, in our series, only two cases displayed a weak
CD38 expression.
In some cases with few PCs, the mathematical model could not
analyze the cell cycle. In most of these cases, this problem could
be overcome by acquiring a larger number of cells or through an
additional acquisition using a live gate for the CD38 strong-positive
cells. Nevertheless, 2% of cases could not be analyzed because their
histograms did not adjust to the mathematical model. Overall, the
percentage of cases successfully analyzed was 91.7% (120/131).
Statistical methods. To estimate the significances ofthe differences between means, Student’s t-test wasused (SPSS, Chicago,
IL). The chi-square test (SPSS) was used for dichotomic variables.
The different clinical and biologic characteristics were considered
individually for their relationship with the probability of achieving
a favorable response to treatment by univariate tests (Student’s ttest, chi-square, correlations and nonparametric tests, SPSS). Subsequently, a multivariate analysis-stepwise regression-(regression,
SPSS)’4 was performed to examine the simultaneous effect of the
different variables on the probability of achieving a favorable response.
The same characteristics were again considered for analysis with
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
SAN MIGUEL ET AL
Fig 1. Image of DNA measurement in a diploid MM. On the left,
the histogram is obtained with the Vindelov et alZJtechnique, showing one GOIG1-phase peak. On the right, image obtained with the
DNAICD38 double-stainingtechnique. The plasma cells are strongly
positive for CD38 allowing its discrimination from the rest of the
normal BM cells for cell-cycle analysis.
respect to their individual and simultaneous effects
on response duration (RD) andoverallsurvival (0S)"univariate and multivariate
analysis, BMDP 1L and 2L, respectively.'' RD and OS curves were
plotted according to the method of Kaplan and Meier, and compared
using the Mantel-Cox, Peto-Prentice, andBreslow tests. Variables
considered for possible inclusion in theregressionanalysiswere
those displayinga significant associationwith survival in the univariateanalysis ( P < .OS) or for which priorstudies had suggested
was
a possibleprognosticvalue.Thestepwiseregressionmethod
discontinued when the P value for entering an additional factor was
above .OS. The model was tested both by including the variables in
a continuous fashion (continuous model)and by grouping them into
categories (dichotomous model).
RESULTS
Cell-cycledistribution.
According to theintensity
of
CD38 expression, myelomatous plasma cells (CD38 strong
positive) were clearly discriminated fromnormalresidual
BM cells, which allowed us to analyze separately the cellcycle distribution of both cell fractions (Fig 1). In the whole
series of 120 newly diagnosed untreated MM patients, it
was observedthat thepercentage of S-phasecells in the
myelomatous fraction was significantly lower than that of
the residual normal BM cell population (3.6% 2 3.1% and
8.0% 2 6.3%, P < ,001). The number of cells in the G2/
M-phase was similar in both fractions, while the proportion
of GO/Gl-phase cells was higher within the myelomatous
PC population (Table 1).
Proliferative activity of BM plasma cells. Myeloma patients were also divided into two groups using the median
value of the present series (3% of S-phase PCs) as the cutoff point according to the proliferative activity of the BM
PCs. The clinical, hematologic and biochemical characteristics of both groups of patients are shown in Table 2. Patients
with more than 3% of S-phase PCs showed a significantly
higher incidence of anemia ( P = .04) and hypercalcemia ( P
= .008); increasedvaluesof
P2M ( P = .001), urea ( P =
.001), and creatinine ( P = .0001) together with higher numbers of PB natural killer cells (0.01), and a poor prognosis
(Table 2, Fig 2) as assessed both by the response to chemotherapy ( P = .02) and survival ( P < .0001) (Table 2). The
median survival of these patients was only 19 months comthan
pared with 56 months for the patients displaying less
3% S-phase PCs. In addition, our results also show that this
cut-off value of S-phase BM PCs allows the discrimination
of two significantly different prognostic groups of patients
even within individual clinical stages (Fig 2, B and C).
S-phase was able to identify high-risk patients that would
have not been recognized according to other criteria. Thus,
within the group of patients with high S-phase PCs (>3%)
46% were under 69 yearsof age, 26% were in clinical stage
I or 11, 42% had few bone lesions (grade 0 or l ) , 27% had
levelsofhemoglobin (Hb) > 10.5 g/dL, 50% hadnormal
albumin serum levels (>3.5 g/dL), 62% hadnormalrenal
function (creatinine < 2.0), 54% did not display hypercalcemia (calcium < 10.5 mg/mL) and 48% hadlowlevels of
P2M ( < 6 pg/mL).
Prolijercltive activity qf residual BM cells. When the patientsweregroupedaccording
to the number of residual
normal BM cells in the S-phase (Table 3), it was observed
that cases displaying a low proliferative activity (<4.5% Sphase cells) showed a significantly higher incidence of severe anemia ( P = .03) and neutropenia ( P = .03) together
with increased numbers of BM PCs ( P = ,003). In addition,
both Bence Jones ( P = ,0002) and diploid ( P = .009) MMs
were more frequent among this group of patients. Interestingly, in six patients,circulating PCswere observed morphologically and five ofthese were in the group of patients
displayinglownumbers of S-phase cells ( P = .01). From
the prognostic point of view, patients in which the normal
residual BM cells displayed a low proliferative rate had a
slightly shorter survival (median survivalof 22 v 36 months),
although the differences didnot reach statistical significance.
Prognostic,factorsandclinicalsystem.
Regarding response to chemotherapy,univariate analysis showed that the
individual disease characteristics associated with a favorable
response (objective and partial remission) were low proliferative activity (<3%) of BM plasma cells ( P = .02), age less
than 69 years ( P = .02), ECOG less than 3 ( P = .002), &M
less than 6 pg/mL ( P = .002), Hb greater than 10.5 g/dL ( P
= .009),creatinine greater than 2 mg/mL ( P = .0l), low
serum urealevels ( P = .03), calcium less than 1 1 mg/mL
( P = .02), and earlyclinical stages (stage I and 11) ( P =
.03). However, the only independent prognostic factors predicting the response to chemotherapy were found to beP2M
serum levels and the performance status ( P < .05). Regarding response duration, the only two variables significantly
Table 1. Cell-Cycle Distribution of Myelomatous Plasma Cells
and Normal Residual BM Cells
Myelomatous Cells
GO/Gl-phase
S-phase
GP/M-phase
W O )
Normal Cells
PValue
94.2 i 4.3 (94.8)
3.6 -t 3.1 (3.0)
2.1 i 2.1 (1.3)
90.4 2 4.6 (90.3)
8.0 i 6.3 (7.5)
2.1 2 2.2 (1.6)
<.001
<.001
,299
Results expressed as mean f SD; median values are included in
brackets.
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
PROGNOSIS IN MULTIPLE MYELOMA
451
Table 2. Clinical Characteristics of MM Patients Accordina t o YO of S-Phase BM Plasma Cells
Characteristic
Age (>69 yrs)
Sex (male)
ECOG 2 3
Presence of plasmacytomas
Advanced clinical stage (stage 111)
7
IgG M component
IgA M component
30%
Bence Jones M component
K MC
71%
h MC
29%
Anemia
< l 0 5 g/dL)
Granulocytes (cells/mm3)
Lymphocytes (cells/mm3)
ESR (mmlhr)
Lactic dehydrogenase (maximum, 320 IU/mL)
Alkaline phosphatase (U/mL)
Total serum proteins (g/dL)
Albumin <3.5 g/dL
Presence of light chain in urine
Urea 250 mg/mL
Creatinine 22.0 mg/mL
Calcium 211.0 mg/mL
P2M2 6 pg/mL
% PC in BM
Presence of PC in PB
PB CD56'/CD3- cells (/mm3)
Response to treatment
Responders (OR PR)
(14%
Failures and early deaths
+
Low Proliferative Activity
(<3% S-phase PC)
High Proliferative Activity
(>3%S-phase PC)
43%
43%
20%
8%
65%
56%
38%
14%
62%
38%
58%
4,000 t 2,300
1,900 t 840
110 t 62
229 t 90
163 t 76
9.5 t 2.7
35%
39%
31%
2%
,008
16%
22%
48 rt 24
4%
300 +. 200
55%
57%
32%
11%
5%
53%
CR)
81%
19%
15%
76%
3,500 t 2,000
2,150 t 1,200
109 t 33
266 t 109
181 t 167
9.2 c 2.0
50%
47%
60%
36%
39%
55%
47 ? 28
5%
775 t 590
56%
44%
PValue
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
.04
NS
NS
NS
,058
NS
NS
. l1
NS
,001
<.00001
.001
NS
NS
.011
.02
.02
Results expressed as mean t SD and percentages.
Abbreviations: NS, not significant ( P > .05); PC, plasma cells;CR. complete response; OR, objective response;PR, partial response; MC,
monoclonal component.
associated with a shorter response duration of independent
value were the presence of more than 3% of S-phase PCs
(P = .002) (Fig 3) and the presence of plasmacytomas at
diagnosis (P = .03). Regarding the population with high S phase, only 44% of patients achieved an OR, and 12% had
a minor response. In the remaining patients, there was no
response. Interestingly, 21% of patients with OR relapsed
within the following 4 months.
The question of whether the proliferative activity of BM
PCs has independent prognostic influence on survival, because of its correlation with other prognostic parameters, was
investigated using the Cox regression analysis. Multivariate
analysis (Table 4) showed that the percentage of S-phase
PCs was the best single parameter for predicting survival in
MM patients; this parameter, together with the P2M serum
levels, performance status and age were the best combination
of disease characteristics for predicting survival in MM. In
10 of 60 patients (17%) with high S-phase PCs (>3%), this
was the only adverse prognostic feature because all the other
independent prognostic factors @,M, age, and performance
status) were favorable. When only patients less than 65 years
old were included in the multivariate analysis, the only disease parameters to emerge as significant prediction were the
percentage of S-phase cells (P< .001) and P2M serum levels
( P = .05).
According to the hazard rate predicted by the regression
model (Fig 4) patients were divided into three risk groups
using cut-off points of 6 and 8.5; a high-risk group including
21% of the patients, an intermediate-risk group including
53%, and a low-risk group including 26% were found. The
median survivals of these three groups were 9 and 36 months
and not yet reached, respectively ( P < .0001). Interestingly,
none of the 31 patients included in the low-risk group have
died so far. All of these patients had symptomatic MM, 54%
of them were in clinical stage I11 and extensive bone lesions
(grade 2 or 3) were present in 46% of these cases. Their
median age was 64. Anemia (<10.5 g/dL) was observed in
SO%, albumin less than 3.5 g/dL in 24% and serum P,M level
between 4 and 6 mg/dL were present in 24%. To simplify the
classification of MM patients into risk groups, an alternative
model was developed according to the scoring indices shown
in Table 5. The survival curves of patients with low risk
(score 0), intermediate risk (score 1, 2 or 3) and high risk
(score 4 or 5 ) were similar to the ones described above, and
the differences between them were also highly significant
(P < .0001) (Fig 5).
DISCUSSION
Accurate prognostic evaluation is mandatory for the selection of optimal therapeutic strategies in hematologic malignancies, especially when a decision on treatments with increased concomitant morbidity and mortality, such as
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
452
SAN MIGUEL ET AL
Table 3. Clinical Characteristics of
MM
Characteristic
Age (yrs)
Sex (male)
ECOG 2 3
Advanced clinical stage (stage 111)
Anemia (Hb d . 5 g/dL)
Lactic dehydrogenase (maximum, 320 IU/mL)
Total serum proteins (g/dL)
Albumin <3.5 g/dL
Urea 2 5 0 mg/mt
Creatinine 22.0 mg/mL
Calcium 211.0 mg/dL
P2M 26 d m L
% PC in BM
Presence of PC in PB
Diploid MM
Response to treatment
Responders (OR PR)
Failures and early deaths
Patients According to % of Normal Residual S-Phase BM Cells
Low Proliferative Activity
(S-phase < 4.5%)
High Proliferative Activity
(S-phase a 4.5%)
69 2 9
42%
28%
68 i 11
.009
53%
240260 i 120
9.2 C 2.9
29%
50%
17%
25%
34%
59% 2 27%
16% (5/34)
59%
+
NS
NS
NS
.l2
.03
NS
NS
NS
NS
NS
NS
NS
,003
.01
54%
30%
67%
25%
i 93
2 2.1
24%
43%
I 8%
29%
32%
42% ? 24%
81%
9.6
P Value
1% 11/86)
33%
62%
38%
60%
40%
NS
NS
Results expressed as mean SD and percentages.
Abbreviation: NS, not significant ( P > .05).
transplantation, must be made. Moreover, a good staging
system would allow the identification of patients at risk of
early death who may benefit from new experimental treatments. The plasma cell labeling index (PCLI), assessed as
the number of plasma cells that incorporate thymidine analogues, has been shown to represent an important prognostic
factor in MM, potentially useful for predicting progression
of the tumor clone.2,3,6,8.9
However, its determination is rela-
p <O.Oool
4L"""
l
_""56 months l
1
tively complex, time consuming, and difficult to standardize.
Because of this, at present, these techniques aimed at assessing cell proliferation are restricted to relatively specialized clinical research laboratories.26In addition, in most MM
patients, the PCLI is quite low; in fact, only 10% of cases
display a PCLI above the cut-off value of 2%' such that
only a minority of patients are identified as high risk because
most patients lie within a tight interval of variation. In recent
years, different flow-cytometry techniques have been developed to assess the proliferative rate of tumor cells based
on their DNA content per cell measured either as a single
parameter or combined with the quantification of the incorporation of the BdrU and IdUr thymidine-analogues, specifically identified with M O A ~ S . Although
~ ' ~ ~ ~ cell-cycle
analysis using flow cytometry has now been performed in
this application for more than 15 years, few clinical studies
have been reported in MM patients." This is probably because the measurement of cell proliferation in total BM is
1
-l
oh
10
20
30
40
80
70
Months fmn dii@?osfs
50
c -"""
26 momha
-
rWIm.Eh*d
M
9O
"I
"""_
il
phaSe f
C 23%
-S
S p h w PC 4%
-" 7
"
"""""
-
J
I
.E
"S"
a
"
ClylJorJ
10
80'
17monh.
"hu
-5ghu23U
IL 20
00
I
111( n = W
U)
0
1
I
I 61months
I
13 months
1
0
02 20 0 9 9 0 0 44
L".
p = 0.001
00
5
50 0 6 6 00 77
0
Months from nvni88ion
Fig 2. Survival curves according to the percentage of S-phase
plasma cells in the whole series (A), early and intermediate clinical
stages (B), and advanced clinical stages (C).
Fig 3. Response duration curves according to the percentage of
S-phase plasma cells in the whole series.
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
453
PROGNOSIS IN MULTIPLEMYELOMA
Table 4. Cox Model Relating Pretreatment Characteristicsto Overall Survival in the Whole Series
Coefficient
Step
SE
Factor
0.5774
PC in S-phase
2 3%
0.3996
P2M 2 6 pglmL
12
ECOG 2 3
yrs
Age 2 69 0.4153
1
2
3
4
0.41
2.2185
0.8117
1.1756
0.9067
P
Log Likelihood
<.001
,002
,012
.022
-111.912
-107.071
-103.951
-101.313
Regression model: Ln[L(t)/l(lo)l= (2.2,s-phase PC) + (0.8.P2M)+ (1.2.ECOG) + (0.9.age). In thisequation, variables are considered as dichotomous, with a value of 1 (if their value was under the cut-off point) or2 (if their value was above the cut-off point).
not an appropriate way of assessing plasma cell proliferation
since normal residual cells are also proliferating in the BM
of myeloma patients. The CD38/propidium iodide doublestaining technique used in the present study" allows the
simultaneous evaluation of the myelomatous population specifically identified as being strongly CD38 positive and the
residual normalBM cell populations. Our data showthat
myelomatous PCs proliferate relatively slowly as compared
with residual BM cells.
So far, very little attention has been paid to the residual
normal BM cell populations in MM and, to the best of our
knowledge, this is the first study in which their proliferative
activity has been correlated with the most relevant clinical
disease characteristics. Our results show that both Hb levels
and the number of PB granulocytes in MM patients are
related to the proliferative activity of residual BM cells as
an index of their regenerative activity.
Regarding the proliferative activity of myelomatous PCs
we observed that the number of S-phase PCs correlated with
several prognostic factors, such as the level of Hb, calcium,
creatinine, and P2M. Moreover, using an S-phase cut-off
value of 3%, it was possible to discriminate two subgroups
of patients with significant different prognoses, even within
individual clinical stages. In a similar way to what has been
reported for the PCL16-*and other chronic lymphoprolifera-
tive disorders,29multivariate analysis showed that the number of S-phase PCs was the most important independent
prognostic factor and that, together with P2M, age, and the
performance status, it represented the best combination of
disease characteristics for stratifying patients according to
prognosis, thus providing a useful basis for risk-directed
therapy. Accordingly, it should be emphasized that together
with a group of patients at risk of early death (22% of our
cases), the model described here identified another similar
group of patients, all of which are alive and are characterized
by a low percentage of S-phase PCs, low serum 0,M levels,
and age less than 69 years. According to classic criteria,
such as the clinical stage (54% of them were in stage 111)
or anemia (50% had Hb < 10.5 g/dL), these patients would
not have been recognized as low-risk cases and, therefore,
might have been subjected to unnecessary, perhaps lifethreatening, aggressive treatments. A similar observation has
been recently reported by Greipp et aI9 using a stratification
model based on PCLI and P2M levels; in this study, eight
of nine patients under the age of 65 with low PCLI and P2M
were alive after more than 5.5 years of follow up.
The percentage of S-phase PCs apparently does not play
a crucial role in predicting response to treatment because it
was not selected in the multivariate analysis for response.
Other studies based on the PCLI have reported similar results2; this maybe because response to treatment seems
mainly to depend on tumor cell mass' and the presence of
mechanisms of drug resi~tance.~'.~'
By contrast, we observed
that response duration was related to the proliferative activity
of the PCs because the number of S-phase cells was the
most important prognostic factor for predicting relapse. This
Table 5. Simplified Prognostic Classification of MM
Variable
L """"
~
I
20.
1
I
IOd
io
-P
Lm ria:
latsnnsdiaterisk
High risk
io
so
io
Global
p < o.ooo1
io
b 0 7 o
io
Monthsm~ls
n
Dad
9
Alive
31
Median survival
31
63
23
20
40
26
36 moatha
9moolhs
6
28) montha
Fig 4. Survival curvesaccording to the prognostic model using
the regression equation: ER = (2.2 S-phase PCs) + (0.8 -&M) + (1.2
ECOG) + (0.9 Age).
S-phase plasma cells
<3%
23%
P2 M serum levels
<6 pg/mL
2 6 pg/mL
Performance status
ECOG 1 3
ECOG 2 3
Age
<69 yrs
269 yrs
Score
0
2
0
1
0
1
0
1
The sum of scores gives three stages: stage I, score 0; stage II: score
lto 3; stage 111, score 4 to 5.
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
454
SAN MIGUEL ET AL
Median not reached
20
I'C
L"-l
l
I
*,(SW 0
g21
-*(@Q
High&(StlSem,
0
n
23
44
21
5
SOm&
36 months
9 months
Fig 5. Survival curves according to the simplified classification of
MM patients. Values acording to the score model shown in Table 5:
S-phase PCs 23% (2). ~ 3 (0);
% &M 2 6 pg/mL (l),<6 pg/mL (0);
ECOG 2 3 (l),<3 (0);age 269 yrs (l),C69 yrs (0).The sum of these
values gives the score risk.
observation suggests that the proliferative activity of the
residual cells resistant to chemotherapy is a major factor for
disease progression.*
In summary, our results show that the assessment of the
number of S-phase PCs at flow cytometry, combining CD38
and PI staining, is the most significant independent predictor
of survival in MM. Moreover, the classification system proposed here includes the three basic features that define the
prognosis in most tumor models: (1) the malignant cells
growth (assessed by plasma cells' S-phase), (2) the tumor
load (assessed by &M), and (3) the patients' ability to tolerate the malignancy and the treatment (age and performance
status). Finally, this classification could also be used for
planning treatment in patients who are younger than 65 years
old, have a good performance status, and who are candidates
for transplantation.
REFERENCES
1. Drewinko B, Alexanian R Growth kinetics plasma cell my-
eloma. J Natl Cancer Inst 57:333, 1977
2 . Durie BGM, Salmon SE, Moon TE: Pretreatment tumor mass,
cell kinetics, and prognostic in multiple myeloma. Blood 55:364,
1980
3. Greipp PR, Kyle RA: Clinical, morphological, and cell kinetic
differences among multiple myeloma, monoclonal gammopathy of
undetermined significance, and smoldering myeloma. Blood 62: 166,
1983
4. Dune BGM, Young LA, Salmon SE: Human myeloma in vitro
colony growth: Interrelationship between drug sensivity, cell kinetics, and patient survival duration. Blood 61:929, 1983
5 . Boccadoro M, Gavarotti P, Fossati G, Pileri A, Marmont F,
Neretto G, Gallamino A, Volta C, Tribalto M, Testa MG: Low
plasma cell 3(H) thymidine incorporation in MGUS, smoldering
myeloma and remission phase myeloma: A reliable indicator of
patients not requiring therapy. Br J Haematol 58:689, 1984
6. Boccadoro M, Massaia M, Dianziani U, Pileri A: Multiple
myeloma: Biological and clinical significance of bonemarrow
plasma cell labeling index. Haematologica (Pavia) 72:171, 1987
7. Greipp PR,Katzmann LA, O'Fallon WM.Kyle RA: Value
of beta-2-microglobulin levels and plasma cell labeling indices as
prognostic factors in patients withnewly diagnosed multiple myeloma. Blood 72:219, 1988
8. Durie BGM, Bataille R: Therapeutic implications for myeloma
staging. Eur J Haematol 51 :l 1 1, 1989 (suppl)
9. Greipp PR, Lust JA, O'Fallon M, Katzmann JA, Witzig TE,
Kyle RA: Plasma cell labeling index and p2-microglobulin predict
survival independent of tymidine kinase and C-reactive protein in
multiple myeloma. Blood 81:3382, 1993
10. HedleyDW, Shankey TV, Wheeless LL: DNA Cytometry
Consensus Conference. Cytometry 14:471,1993
I I . Shankey TV, Ravinovitch PS, Bagwell B, Bauer KD, Duque
RE, Hedley DW, Mayall BH, Wheeless LL: Guidelines for implementation of clinical DNA cytometry. Cytometry 14:472, 1993
12. Duque RE, Andreeff M, Braylan RC, Diamond LW, Peiper
SC: Consensus review of the clinical utility of DNA flow cytometry
in neoplastic hematopathology. Cytometry 14:492, 1993
13. Smith L, Barlogie B, Alexanian R: Biclonal and hypodiploid
multiple myeloma: Am J Med 80:841, 1986
14. Shimazaki C, Gotoh H, Oku N, Ashihara E, lnaba T, MuraM, Fujita N: Noveltype of
kami S, Ura Y, ItohK,Nakagawa
clonally involved cytoplasmic immunoglobulin-negative cells in
multiple myeloma: Flow cytometric study. Acta Haematol 88:56,
I992
15. Terstappen LWMM, Johansen W, Segers-Nolten IMJ, Loken
MR: Identification and characterization of normal human plasma
cells in normal human bone marrowby high resolution flow cytometry. Blood 76:1739, 1990
16. Harada H, Kawano MM, Huang N, Harada Y, Iwato K, Tanabe 0, Tanaka H, Sakai A, Asaoku H, Kuramoto A: Phenotypic
difference of normal plasma cell from mature myeloma cells. Blood
81:2658, 1993
17. San Miguel JF, Garcia-Sanz R, Gonzzilez M, Moro MJ, Hernandez JM, Ortega F, Sanz MA, Borrego D, Carnero M, Casanova
F, Jimenez R, Porter0 JA, OrfZo A: Estudios inmunofenotipicos y
de contenido de ADN en Mieloma Multiple: lmplicaciones clinicas.
Sangre 38:87, 1993 (suppl)
18. Orfgo A, Garcia-Sanz R, Lopez-Berges MC, Vidriales MB,
Gonzalez M, San Miguel JF: A newmethod for the analysis of
plasma cell DNA content in multiple myeloma samples using a
CD38/propidium iodide double staining technique. Cytometry (in
press)
19. Costa A, DelBin0 G, HammouA, Porschen R,VanHove
L, Valangendonck F, Knape WA, Bach BA: A multicenter study of
reproducibilty of FCM cell cycle estimates. Cytometry 5:633B, 1991
(suppl, abstr)
20. Chronic Leukemia-Myeloma Task Force, National Cancer Institute: Proposed guidelines for protocol studies: 11: Plasma ceil myeloma. Cancer Chemother Rep 4:145, 1973
2 I . Blade J, San Miguel JF, Aka16 A, Maldonado J, Sanz MA,
Garcia-Conde J, Moro MJ, Alonso C, Besalduch J, Zubizarreta A,
Besses C, Gonzilez Brito G , HemAndez-Martin J, Femindez-Calvo
J, Rubio D, Ortega F, Jimbnez R, Colominas P, Faura MV, Font L,
Tortosa J, Doming0 A, Fontanillas M, Rozman C, Estap6 C: Alternating combination VCMPNBAP chemotherapy versus melphalad
prednisone inthe treatment of multiple myeloma: A randomized
multicentric study of 487 patients. J Clin Oncol 1 I: 1165, 1993
22. Durie BGM, Salmon SE:A clinical staging system for multiple myeloma. Correlation of measured myeloma cell mass with presenting clinical features, response to treatment and survival. Cancer
36:842, 1975
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
PROGNOSIS IN MULTIPLE MYELOMA
23. Vindelov LL, Christehnsen IBJ, Jensen G, Nissen NI: Limits
of detection of nuclear DNA anbnormalities by flow cytometric DNa
analysis: Results obtained by a set of methods for sample storage,
staining and internal standardization. Cytometry 3:332, 1983
24. Cox D R Regression models and life tables: J R Stat SOC(B)
34:187, 1972
25. Dixon WJ: (Biomedical-Data Package) Statistical software.
University of California, Berkeley, CA, 1983
26. Boccadoro M, Gallone G , Frieri R, Pileri A, Bataille R, Klein
B: Plasma cell labeling index, &microglobulin and C-reactive protein: What is the best combination for myeloma prognosis? Blood
82:3507, 1993
27. Wilson GD, McNally NJ: Measurement of cell proliferation
using bromodeoxiuridine, in Hall PA, Levison DA, Wright NA (eds):
Assessment of Cell Proliferation in Clinical Practice. London,UK,
Springer-Verlag, 1992, p 113
A, Gonzfilez M, Ciudad J, L6pez-Berges MC, L6pez
28. M O
455
A, Vidriales B, Macedo A, San Miguel JF: Cell cycle andDNA
aneuploidy: Biological bases and terminology, in Sanpedro A, O f i o
A (eds): DNA Cytometric Analysis. Oviedo, Spain, Universidad de
Oviedo, 1993, p13
29. W a o A, Ciudad J, Gonzilez M, San Miguel JF, Garcia AR,
Mpez-Berges MC, Ramos F, Del Caiiizo MC, Rios A, Sanz M,
L6pez-Borrasca A: Prognostic value of S-phase white blood
cell count in B-cell chronic lymphocitic leukemia. Leukemia 6:47,
1992
30. Dalton WS, Durie BGM, Alberts DS, Gerlach JH, Cress AE:
Characterization of a new drug-resisteant human myeloma cell line
that expresses P-glycoprotein. Cancer Res 46:5125, 1986
31. Dalton WS, Grogan TM, Rybski JA, Scheper JR, Richter L,
Kailey J, Broxterman HJ, Pinedo HM, Salmon SE: Immunohistochemical detection and quantitation of p-glycoprotein in multiple
drug resistence human myeloma cells: Association with level of drug
resistence and drug accumulation. Blood 73:747, 1989
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
1995 85: 448-455
A new staging system for multiple myeloma based on the number of
S- phase plasma cells
JF San Miguel, R Garcia-Sanz, M Gonzalez, MJ Moro, JM Hernandez, F Ortega, D Borrego, M
Carnero, F Casanova and R Jimenez
Updated information and services can be found at:
http://www.bloodjournal.org/content/85/2/448.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.