From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
Direct Relationship Between Remission Duration in Acute Myeloid Leukemia and
Cell Cycle Kinetics: A Leukemia Intergroup Study
By Azra Raza, Harvey D. Preisler, Roger Day, Zahida Yasin, Mike White, Joe Lykins, Cathi Kukla, Maurice Barcos,
John Bennett, George Browman, Jack Goldberg, Hans Grunwald, Richard Larson, J a m e s Vardiman, and Ralph Vogler
Cell cycle characteristics including labeling indices (LI),
duration of S-phase (Ts), and total cell cycle time (Tc) were
determined in 64 standard-risk, newly diagnosed patients
with acute myeloid leukemia following an infusion of
bromodeoxyuridine. Remission induction therapy consisting of cytosine arabinoside and daunomycin was then
administered to all patients, followed by three courses of
consolidation to those who achieved complete remissions
(CR). Older patients appeared to have more rapidly cycling
cells ( P = .003). No unique cell cycle characteristics were
identified for patients who achieved remission versus
those who had resistant disease. However, the pretherapy
cell cycle characteristics were a strong prognosticator for
remission duration. CR patients were divided into those
whose leukemic cell Tc were above median (A) and below
median (B). Among 14 B patients, median duration of
responsewas 21 1 days, and all relapsed by day 600. Among
18 A patients, the median has not as yet been reached,
with nine patients in continuous complete remission (log
rank P = .007, Wilcoxon P = .04). W e conclude that cell
cycle characteristics of leukemic cells play a role in determining remission duration, perhaps because the leukemic
cells of the former patients regrow slowly between courses
of chemotherapy.
0 1990 by The American Society of Hematology.
D
Estimation of the percentage of S-phase cells. After a diagnosis
of standard-risk AML was established in every patient, BrdU was
administered at a dose of 100 mg/m2 in 50 mL of 0.9% normal saline
IV over 1 hour. Immediately at the end of the infusion, peripheral
blood (PB) bone marrow aspirate (BMasp) and biopsy samples were
obtained from every patient. The biopsy samples obtained at the end
of BrdU infusion were fixed in Bouin's solution and processed in
plastic by using glycol methacrylate (GMA) as previously described.'
Sections were then processed by the monoclonal anti-BrdU antibody
as described before: counterstained with hematoxylin and MayGriinwald stains, and at least 2,000 cells were counted from each
biopsy section from at least five different fields by a single observer.
Determination of Ts. The post-BrdU infusion BMasp was doublelabeled with 'HTdr as described before? At least 2,000 positively
labeled cells were counted per slide by a single observer. A cell was
considered positively labeled for BrdU incorporation if there was any
fluorescence detected over its nucleus, and positive for 'HTdr
incorporation if at least five grains (if the background was less than 1
grain/cell, then three grains) were detected on its nuclear surface.
The cells considered positive for double labeling demonstrated both
fluorescence and silver grains over their nuclei." Ts was then
calculated by the following formula as described by Wimber and
EFINING the proliferative characteristics of leukemic
cells has been the subject of extensive investigation
and controversy for several decades.'.* While it is intuitively
accepted that the rate of division of leukemic cells should be a
major determinant of both the course of leukemia and its
response to chemotherapy, this proposition has not been
clearly demonstrated. In fact, there is an abundance of
contradictory reports in the
Perhaps the single
most important reason for this contradiction has been the
absence of an easily reproducible method that would permit
the assessment of not only the percentage of cells in S-phase,
but also the more important cell cycle parameters such as the
duration of the cell cycle and of S-phase.
The major impediment to developing such a method has
been the absence of two easily distinguishable probes that
would be incorporated into the DNA. We were able to
overcome this problem by introducing bromodeoxyuridine
(BrdU) into DNA as the first label in vivo, and tritiated
thymidine ('HTdr) as the second D N A label in vitro.
Combining immunofluorescent detection of BrdU (monoclonal anti-BrdU antibody) and autoradiography allowed us
to measure the influx and eflux of cells into and out of
S-phase, thereby providing an estimate of the duration of
S-phase (Ts) and hence calculation of total cell cycle times
(Tc).' We have used this methodology to assess the relationship of cell cycle parameters to treatment outcome in 54
newly diagnosed, uniformly treated patients with acute
myeloid leukemia (AML) and have found a highly significant relationship with remission duration.
MATERIALS AND METHODS
Fifty-four standard-risk, newly diagnosed AML patients received
BrdU intravenously (IV) before starting remission induction therapy. The BrdU protocol was reviewed and approved by the Institutional Review Board (IRB) of each participating member institution, as well as by the National Cancer Institute (NCI) and the Food
and Drug Administration (FDA). The BrdU used in this study
(Protocol INT0050) was supplied by the NCI. Informed consent as
per guidelines required by the local IRB, as well as the NCI and
FDA, was obtained from every patient before the administration of
BrdU. The following sections describe the methodology involved in
greater detail.
Blood, Vol76, No 1 1 (December 1I, 1990: pp 2 19 1-2 197
From the Roswell Park Memorial Institute, Buffalo, N K Barrett
Cancer Center, University of Cincinnati, Cincinnati, OH; Pittsburgh
Cancer Institute, Pittsburgh, PA; University of Rochester Cancer
Center, Rochester, NY; Hamilton Civic Hospital, Hamilton, Ontario, Canada; Health Science Center at Syracuse, Syracuse, N K
Long Island Jewish Hillside Medical Center, Jamaica, NY; University of Chicago Medical Center, Chicago. IL; and Emory University,
Atlanta, GA.
Submitted May 25, 1990; accepted August 6, 1990.
Supported in part by Public Health Service Grant Nos. CA41285
and CA28734.
Address reprint requests to Azra Raza. MD, Associate Professor
of Medicine. Director of Hematologic Oncology Program, University of Cincinnati Medical Center, 231 Bethesda Ave (ML 508).
Cincinnati, OH 45267.
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 1990 by The American Society of Hematology.
0006-4971/90/7611-0013$3.00/0
2191
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
2192
Quastler”: Ts = (DL + ’HTdr) x t/BrdU, where DL is the
number of cells demonstrating double labeling, ’HTdr is the number
of cells only showing ’HTdr incorporation, BrdU is the number of
cells revealing only BrdU incorporation, and t is the time interval
between the two labels. The time interval t in this formula was taken
as 1 hour because that was the time lapse between the start of the
first and the start of the second label.
Determination of Tc. The length of Tc was determined by using
the formula described beforeI2as follows: Tc = Ts x GF/LI, where
G F is the growth fraction and LI is the percentage of S-phase cells
obtained from plastic embedded biopsy samples. Because we were
calculating the doubling times of only those cells actively engaged in
cycle, the G F was assumed to be 100%in each case.
Mechanics of sample transport. The BrdU protocol INT-0050
is a Leukemia Intergroup study, providing for administration of the
in vivo label locally at each member institution listed. All post-BrdU
samples were processed initially at local institutions. For example,
post-BrdU PB and BM cells were density-cut and double-labeled
with ’HTdr, placed on coverslips, and fixed in 70% ethanol. These
coverslips can be stored permanently in 70% ethanol. The BM biopsy
samples were fixed and dehydrated locally and then transported to
Roswell Park Memorial Institute (RPMI; Buffalo, NY) within a
week. The PB and BMasp samples were transported regularly to
RPMI. All monoclonal antibody (MoAb) applications, autoradiographic procedure, as well as GMA infiltration of biopsies and their
processing were performed centrally at RMPI in the laboratory of
Dr Azra Raza. All slides were interpreted by at least two individuals
(Dr Raza was one of them in every case) who blindly counted each
others’ slides. The majority of samples were assigned a coding
number and individuals interpreting slides did not know the names of
patients, the institution they were from, or outcome of therapy in any
of the cases.
Clinical data. Fifty-four standard-risk, newly diagnosed AML
patients are the subject of this report. The term standard risk refers
to all previously untreated patients with AML (FAB, MI, M2, M3,
M4, M5, M6, M7). Patients in any of the groups listed below are
considered to be high-risk patients and are treated on a different
chemotherapy regimen. They will be the subject of a subsequent
report. (1) Patients who were older than 70 years of age. (2) Patients
with overt congestive heart failure and patients with uncontrollable
ventricular arrythmias, both of which precluded such individuals
from receiving anthracyclines. (3) Patients whose performance
status was 4 (completely bedridden). (4) Patients with AML
secondary to chemotherapy, radiation therapy, exposure to toxic
agents, or patients with a history of myelodysplastic syndrome.
Induction therapy consisted of daunomycin (DNR) 50 mg/m2/d (35
mg/m2/d for patients older than 60 years) by continuous IV infusion
on days 1,2, and 3, and cytosine arabinoside (araC) at 100 mg/m2/d
by continuous IV infusion on days 1 through 7 (7 + 3). Patients
whose BMasp contained 210% abnormal cells on day 6 were given 3
additional days of araC (10 + 3). Complete remission (CR) was
defined according to the Cancer and Leukemia Group B (CALGB)
criteria described before.13 The treatment failures were classified
according to our previously described classification system, which
provides for recognizing those individuals who failed therapy because of persistent leukemia (resistant disease or RD) and those
patients who died during induction therapy or during the hypoplastic
phase induced by chemotherapy other^").'^ The CR rate on this
protocol (which is being performed at the present time) is 76%, with
15% patients failing due to RD and 8% dying during induction
therapy. Following achievement of CR, all patients were administered three courses of intensive consolidation therapy. Courses 1 and
3 consisted of 7 + 3 (daunomycin 50 mg/m2/d for 3 days given by
continuous infusion and araC 100 mg/m2/d x 7 days). Course no. 2
RAZA ET AL
consisted of araC at 3 Gm/m2 IV every 12 hours for 4 days (reduced
to 2 Gm/m2 for patients older than 60 years).
Statistical methods. Clinical data were collected on INT-0066
locally by each participating institution and sent for entry into the
Leukemia Intergroup (LIG) data collection’s office by Frontier
Science, Buffalo, NY. The laboratory data were entered onto forms
and sent for decoding of samples to the LIG operations office and
then entered into the computer. All statistical tests performed were
two-sided. All correlations presented herein are Spearman correlations. The statistical significance of correlations were tested with
permutation tests. Relationships with clinical outcome were tested
with Fisher exact tests. Two types of analyses were performed to
relate cell cycle variables to remission duration. One type classified
patients as high or low on each variable, and compared the resultant
groups using log-rank tests and Gehan-Gilbert-Wilcoxon tests (hereafter referred to as Wilcoxan tests). In each case, “high” meant that
the value in question was higher than the median among the 54 study
patient values. The other type of analysis was to rank each variable
and fit accelerated failure time models’5 linear in the ranks. The
distributions of remission durations were assumed to be Weibull
distributions. The statistical analysis systems (SAS) procedures
LIFETEST and LIFEREG were used for these analyses. Two
patients for whom the duration of response was not available are
excluded from these analyses (patient no. 1411 in Table 1 is thereby
excluded in Ts analysis, but Tc is unavailable for this patient. The
other patient who was excluded is no. 1215).
RESULTS
Measurements of cell cycle parameters in vivo. Fifty-four
standard-risk, newly diagnosed A M L patients were studied.
Figures 1A and B show post-BrdU infusion BMasp a n d
biopsy sections from a n A M L patient, treated with t h e
anti-BrdU antibody followed by a secondary antimouse
antibody conjugated t o peroxidase. Reddish-brown staining
is clearly visible over the nuclei of S-phase cells. Table 1 gives
details of t h e cell cycle parameters for all 54 patients. A s c a n
be seen from this table, the percentage of S-phase cells or t h e
LI was available from BMasp in 49 patients and from B M
biopsy in 50 patients. T h e mean LI from BMasp was 11.1%
(median = 9%, range 2% to 40%). while t h e mean LI from
BM biopsies was 27.3% (median = 27.50%, range 9% to
47%). These differences in LI between aspirates a n d biopsies
can be most readily explained by the variable hemodilution
of the aspirate, which often occurs.’6-’8
Measurements of t h e Ts a r e available for 48 patients
(Table 1). The mean Ts of the group was 14.7 hours
(median = 13.5 hours, range 6 to 43 hours). Tc measurements are available in 45 patients. The mean Tc was 59.8
hours with a median of 48 hours and range of 18 to 21 1
hours.
Relationship of cell cycle parameters to patient and
leukemic cell characteristics. A n assessment was made to
determine whether any relationships existed between t h e cell
cycle parameters and t h e characteristics of t h e patients or
their leukemic cells. Among t h e 5 4 patients, 26 were male
and 28 female. Table 2 shows t h e relationship between t h e
four cell cycle kinetic parameters and age a n d sex in these
patients. No relationship was discernable between t h e cell
cycle parameters a n d sex, but there was a statistically
significant relationship with age. Increasing age was associ-
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
2193
CELL KINETICS IN AML
Table 1. Cell Cycle Characteristics and Clinical Data on Newly Diagnosed, Standard-Risk AML Patients Treated on INT-0066
SNo.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
Mean
Median
Range
UPN
%LI Asp
%LI Bx
Ts (h)
Tc (h)
Tx Outcome
1175
1195
1200
1201
1205
1209
1212
1215
1217
1225
1226
1227
1229
1231
1242
1259
1276
1277
1286
1296
1305
1306
1312
1319
1321
1339
1354
1366
1368
1373
1375
1376
1378
1381
1385
1388
1393
1398
1401
1402
1403
1404
1409
141 1
1413
1416
1421
1427
1429
1438
1445
1446
1455
1461
2
14
8
21
24
9
33
24
18
21
12
9
19
57
38
21 1
CR (2)
CR
RD
CR
0th F
CR
RD
CR
CR (2)
CR
3
8
7
7
5
8
8
3
10
5
8
24
5
14
-
18
15
13
5
17
23
14
15
20
16
8
10
15
17
14
5
13
9
10
3
40
8
5
21
24
10
12
6
4
7
3
5
11.10
9.00
2-40
18
24
33
25
10
23
28
21
36
30
25
35
24
38
39
28
34
26
35
28
32
26
23
27
9
7
22
15
15
12
10
6
8
31
14
43
12
16
-
13
14
21
7
16
23
13
8
50
33
-
83
63
52
40
60
35
111
67
119
40
64
54
39
54
25
47
88
37
29
-
-
12
22
18
18
8
11
7
19
11
6
19
11
46
96
67
-
-
28
42
24
190
38
18
63
52
31
30
38
47
25
40
31
25
31
33
14
12
20
18
15
15
106
48
32
42
72
38
48
29
26
29
10
29
34
31
21
14
24
27.26
27.50
9-47
-
14
9
8
10
14.69
13.50
6-43
-
-
46
59
42
59.84
48.00
18-21 1
CR (2)
CR
CR
CR
CR
0th F
CR
CR
CR
CR
CR
CR
CR
CR
RD
CR (2)
RD
CR
CR (2)
CR
CR
CR (2)
lncv
0th F
RD
CR
CR
CR
CR
RD
CR
CR
CR
CR
CR
CR
0th F
RD
CR
CR
RD
CR
RD
CR (2)
Abbreviations: %LI Asp, percentage of Ts cells from BMasp; %LI 8x. percentage of Ts cells from bone marrow biopsy; Tx Outcome, treatment
outcome; UPN, unique patient number.
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
RAZA ET AL
2194
time as shown in Fig 2. The data demonstrate that patients
with rapidly cycling cells and patients with slowly cycling
cells can achieve CRs when treated with the remission
induction regimen used in this study.
Relationship of cell cycle parameters and duration of
remission. To determine whether there was any relationship between the pretherapy cell cycle characteristics of
leukemic cells and the duration of remission, the median
value of each cell cycle parameter was used to divide patients
into a slowly proliferative group (above median value [A]
and a rapidly proliferative group (below median [B]). These
median values were then plotted on a survival function curve.
A highly significant relationship between remission duration
and cell cycle time was identified (Fig 3). All of the patients
in long-term remission had pretherapy leukemia cell cycle
times which were greater than the median value. Nine of the
18 patients in this group continue to be in remission. In
contrast, all 14 remission patients whose leukemic cell cycle
times were less than the overall median value relapsed by day
600 with a median duration of remission of 210 days. These
differences are statistically significant (log rank P = .007,
Wilcoxon P = .042). Hence, only patients whose leukemic
cells at the time of diagnosis were slowly cycling have long
remissions. Interestingly, of the nine patients on curve A in
Fig 3, two had acute promyelocytic leukemia (FAB M3) and
the longest survival is an M3 patient.
The Ts was also related to remission duration, but was not
statistically significant (log rank P = .lo, Wilcoxon P = .29)
while neither the pretherapy LI of the BMasp or biopsy were
related to remission duration.
Fig 1. (A) BMasp obtained from an AML patient immediately at
the end of a 1-hour infusion of BrdU and treated with anti-BrdU
antibody shows marked brown staining of cells actively engaged in
DNA synthesis. (B) Immediate post-BrdU infusion bone marrow
biopsy section treated with the anti-BrdU antibody shows a
number of blasts synthesizing DNA. A megakaryocyte is also in Ts.
ated with a shortening of the Tc of the leukemic cells (Tc v
age: r = -.42, P = .003). Similarly, the Ts was inversely
related to patient age ( r = - .27, P = .05).
The French-American-British (FAB) classification of the
leukemia of every patient was available. Table 3 shows the
Tc versus FAB categories. Furthermore, patients were divided into two groups, ie, those having above (A) versus
below (B) median Tc values. Table 3 shows the distribution
of each FAB category into these two groups as well. While no
statistically significant differences were noted, it is interesting to note that all four FAB M 3 patients had above median
Tc values.
Relationship between cell cycle parameters and outcome
of therapy. All patients received araC/DNR remission
induction therapy following the BrdU infusion. The C R rate
for all patients treated on this protocol has consistently been
between 70% to 80%. For the 54 patients being reported here
in whom cell cycle studies were performed, there were 40
CRs, 9 R D failures, and 5 patients died early in therapy or
during the period of marrow hypoplasia. None of the cell
cycle variables examined showed any statistically significant
relationship to induction outcome, including the doubling
DISCUSSION
Two conditions must be satisfied to determine the clinical
relevance of the cell cycle characteristics of malignant
diseases. First, a reliable quantitative measurement of the
proliferative characteristics of the tumor should be available.
Assessment of only the percent of cells in S-phase is not
adequate. Second, the relationship of these parameters to
both the outcome of therapy and the course of the disease
should be sought in a group of patients who have a relatively
comparable disease and who receive the same treatment. In
the present report, we have accomplished both goals by
measuring, in addition to the LI, the Ts and the Tc of the
malignant cells in a uniform group of patients who received
the same therapy. Before discussing the clinical relevance of
cell cycle parameters in AML, it is appropriate to briefly
discuss some aspects of the methodology used in our studies.
(However, this is not a method report, and indeed, detailed
descriptions of the double-label technique have been dealt
with extensively in the past. For detailed answers, please see
references 7,9, and 10).
One question that may be raised by those not familiar with
the methods used in these studies is whether we have
measured the characteristics of leukemic cells or of residual
normal cells in the bone marrow. In reality, this question asks
whether measurements of normal residual stem cells and
normal “differentiating” cells could have affected the assessment of leukemic cell cycle parameters. Normal “stem” cells
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
CELL KINETICS IN AML
2195
Table 2. Cell Cycle Characteristics of Myeloblasts Versus Clinical Characteristics of Standard-Risk, Newly Diagnosed Patients Being
Treated on INT-0066
Cell Cycle Variables
Clinical Variables
Hemoglobin
WBC Count
% Abnormal cells (blood)
Platelet count
Biopsy cellularity
Leukemic cell mass
% Abnormal cells
(bone marrow)
% Abnormal cells
day 6 (bone marrow)
Normalized LDH
Fibrinogen level
Albumin level
LI ASD
LI Bx
Ts
Tc
- .02
.28
.04
0.17
0.22
0.14
0.32
-0.12
0.38
-0.23
0.10
0.22
0.10
-0.05
0.73
0.1 1
0.47
0.09
0.50
0.01
0.93
-0.001
0.99
-0.00
0.99
0.17
0.23
-.27
.06
-0.03
0.79
-0.05
0.7 1
0.0 1
0.92
0.02
0.87
0.25
0.08
-0.02
0.86
0.12
0.42
0.20
0.16
0.03
0.83
0.04
0.79
-0.19
0.19
0.28
0.05
- .42
.86
0.23
0.10
0.05
0.68
-0.12
0.39
-0.33
0.01
0.32
0.02
0.008
0.95
-0.006
0.96
-0.03
0.82
-0.23
0.13
0.1 1
0.42
-0.05
0.69
0.07
0.62
,003
-0.20
0.16
-0.05
0.72
0.08
0.59
0.17
0.26
0.13
0.37
-0.07
0.65
-0.06
0.70
0.07
0.61
-0.02
0.86
-0.08
0.59
-0.08
0.58
0.13
0.37
Leukemic cell mass is the percentage of leukemic cells in marrow aspirate times biopsy cellularity. Normalized LDH is the patient’s LDH level divided by
the highest “normal” value at that institution (laboratory).
represent less than 0.01% of normal bone marrows, and in
leukemic marrows their frequency is presumed to be even
lower. Because the percentage of S-phase cells labeled by
BrdU in the biopsy sections of leukemic marrows is generally
20% or more (see Table I), even if the normal progenitor
cells had an LI of 50%, they would at most represent only
1/1,00Oth of the labeled cells. This small number could
hardly affect the cell cycle calculations. As far as the
question of whether “maturing” normal cells were confused
with leukemic blasts is concerned, the labeling technique
used insures against this. Visual evaluation of every cell that
is being examined permits us to distinguish very clearly
between leukemic blast cells and more mature cells such as
myelocytes, red blood cell precursors, megakaryocytes, and
so on (see Fig 1). Thus, the fact that cell cycle studies are
addressed to floridly leukemic marrows before any therapy,
together with the ability to visually distinguish between blast
cells and the few mature residual normal cells, precludes any
confusion between leukemic versus normal cells.
A second important point to be noted in these studies is the
fact that by labeling S-phase cells in vivo we are able to
measure the BM biopsy LI rather than being dependent on
an aspirate LI, which can often be spuriously low due to
hemodilution. Without this precise measurement of the
percentage of S-phase cells, calculations of Tc cannot be
made accurately. No previous study in the literature besides
our own has used the biopsy LI to calculate Tc.
Finally, our assumption that GF was 100% was based
partly on the fact that there is no reliable method of
estimating this value accurately, and partly on the basis of
Table 3. Tc of Myeloblasts Versus F A 6 Classification in Patients Treated on INT-0066
FAB
n
Mean Tc (h)
Median Tc (h)
M1
M2
M3
M4
M5
M6
7
10
4
14
5
5
70.14
52.3
98.7
55.5
53.6
47.6
52.0
41.0
77.5
51.0
47.0
46.0
Abbreviations: A, number above median Tc; B, number below median Tc.
A
B
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
RAZA ET AL
2196
COMPLETE REMISSION RESBTANT DISEASE
OTHER FAILURE
Fig 2. Duration of cell cycle time versus outcome of remission
induction therapy on INT 0066.
three previous st~dies'~'''that have independently reported
that the noncycling or Go compartment of cells is relatively
small. Clarkson et aI,l9 for example, administered 3HTdr by
continuous infusion for 8 to 10 days and showed that 92% to
98% of bone marrow cells were labeled at the end of this
period. This left a small, but definite proportion of 2% to 8%
noncycling cells in the Go compartment. However, it is
equally possible that these 2% to 8% cells were not truly Go
cells, but only very slowly cycling. Even if DNA-specific
labels can be infused for prolonged periods to leukemic
patients, data would still be unreliable as many Go cells may
be recruited into cycle during this period. More recently, we
have attempted to determine the G F by using MoAbs against
I
1 .o
T
i
8-B
I
8-8
t
A
'
0
:
0
:
5
:
8
A =Above Median
B=Eclorr Median
pz.007
d.I
0.4
;
0.2
+
B--B
M4'
M4'
M i
Fig 3. Relationship between Tc (split into two groups: A,
above median cell cycle time; B. below median cell cycle time) and
remission duration in AML patients treated on INT 0066. Note FAB
categorization of all long-term survivors on curve A.
cycle-specific protein^.'^.'^ However, no independant means
are presently available to confirm these data. In the final
analysis, estimations of biologic parameters such as G F are
validated only if they demonstrate substantial correlations
with clinical course of the disease or outcome of therapy. In
this respect, because some of the parameters, especially those
of Tc, are significantly related to clinical course of AML in
our study, it appears that even if an error is introduced by our
method of calculations it is likely to be a small one.
Having addressed these potential methodologic questions,
let us now review the data that have been presented in this
report. Not surprisingly, as we have previously reported. even
in this relatively homogeneous group of newly diagnosed,
standard-risk AML patients, the cell cycle characteristics of
myeloblasts vary greatly from patient to patient (Table 1). It
is of interest that the leukemic cells of older individuals cycle
more rapidly than those of younger patients. This observation may in part account for the poorer prognosis of older
individuals. Also interesting is the suggestion that the leukemic cells of patients with FAB M 3 have longer cell cycle
times than the leukemic cells of other FAB categories (Table
3), which is perhaps a reflection of the advanced state of
maturation of these cells. In other studies we demonstrated
that the cell cycle times of HL60 cells increase as they
mature.24Indeed, the fact that these cells are already more
mature may help explain why patients with FAB M 3 AML
have such excellent clinical responses to agents that induce
differentiati~n.~~
N o unique cell cycle characteristics helped distinguish
between those patients who achieved C R versus those who
failed remission induction therapy. Because of the high C R
rate of 76% on the treatment protocol INT0066, very few
treatment failures are available for comparison (33 CRs v 9
RDs). Perhaps with greater numbers of cases being studied,
we will be able to identify a relationship between cell cycle
time and remission induction outcome. In contrast, we have
identified a strong relationship between cell cycle time and
remission duration, with the only patients remaining in
continuous remission being those whose leukemic cells proliferate slowly.
Relapse in an AML patient who initially achieved C R
must be a complex phenomenon being related to the amount
of residual disease, the proliferative potential of the surviving
leukemic cells, the rate of regrowth of these cells, the effect of
consolidation therapy on cell numbers, and their level of
differentiation. The lack of relationships between cell cycle
time and remission induction outcome, and the strong
relationship of cell cycle time and remission duration seems,
at first glance, to be paradoxical. However, it becomes
understandable if one interprets the remission induction
outcome data to indicate that when araC/DNR is administered, the killing of leukemic cells is independent of cell cycle
parameters. If this is the case then the most likely explanation for the observed relationship between remission duration
and leukemic cell cycle times is that the cell cycle time plays
a major role in determining the rate of leukemic cell
regrowth between courses of consolidation therapy and
subsequent to the cessation of all therapy. If this hypothesis is
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
CELL KINETICS IN AML
2197
correct, then major efforts should be directed at reducing the
rate Of regrowth Of leukemic
between
Of
therapy. Administration of retinoic acid, a-interferon, etc
between courses of consolidation therapy may be useful in
this regard.
ACKNOWLEDGMENT
The authors thank Angela Cuppone for excellent secretarial
assistance, and Joe Evola for patient illustrations. All laboratory
studies were conducted in Dr Raza’s laboratory at Roswell Park
Memorial Institute, Buffalo, NY.
REFERENCES
1. Clarkson BD, Fried J: Changing concepts of treatment in acute
leukemia. Med Clin North Am 55561, 1971
2. Mauer AM: Cell kinetics and practical consequences for
therapy of acute leukemia. N Engl J Med 293:389, 1975
3. Andreeff M: Cell kinetics of leukemia. Semin Haematol
23:300, 1986
4. Burke PJ, Owens AH: Attempted recruitment of leukemic
myeloblasts to proliferative activity by sequential drug therapy.
Cancer 28:830,1971
5. Crowther D, Beard MEJ, Bateman CJT, Sewell RL: Factors
influencing prognosis in adults with acute myelogenous leukemia. Br
J Cancer 32:456, 1975
6. Hart JS, George SL, Frei E 111, Bodey GP, Nickerson RC,
Freireich EJ: Prognostic significance of pretreatment proliferative
activity in adult acute leukemia. Cancer 39:1603, 1977
7. Raza A, Maheshwari Y, Yasin Z, Mandava N, Mayers GL,
Preisler HD: A new method for studying cell cycle characteristics in
ANLL using double-labeling with BrdU and ’HTdr. Leuk Res
11:1079, 1987
8. Leduc EH, Bernhard W: Recent modifications of the glycol
methacrylate embedding procedure. J Ultrastruct Res 19:196, 1967
9. Raza A, Maheshwari Y, Preisler HD: Differences in cell cycle
characteristics amongst patients with acute nonlymphocytic leukemia. Blood 69:1647, 1987
10. Raza A, Maheshwari Y, Brereton W, Preisler H D An
analysis of cell cycle characteristics and course of disease in ANLL.
Am J Hematol24:65, 1987
11. Wimber DE, Quastler H: A I4C- and ’H-thymidine doublelabeling technique in the study of cell proliferation in tradescantia
root tips. Exp Cell Res 308,1963
12. Quastler H: The analysis of cell proliferation kinetics, in
Lemerton LF, Fry RJM (eds): Cell Proliferation. New York, NY,
Blackwell, 1963, p 18
13. Ohnuma T, Rosner F, Levy RN, Cuttner J, Moon JH, Silver
RT, Blom J, Falkson G, Burningham R, Glidewell 0, Holland J F
Treatment of acute leukemia with L-Asparaginase. Cancer
Chemother Rep 55:269,1971
14. Preisler HD: Failure of remission induction in acute myelocytic leukemia. Med Pediatr Oncol4:275, 1978
15. Kalbfleisch JD, Prentice RL: The Statistical Analysis of
Failure Time Data. New York, NY, Wiley, 1980
16. Dosik GM, Barlogie B, Gohde W, Johnston D, Terell JL,
Drewinko B: Flow cytometry of DNA content in human bone
marrow. A critical reappraisal. Blood 55:734, 1980
17. Hiddemann W, Buchner T, Andreef M, Wormann B,
Melamed M, Clarkson BD: Bone marrow biopsy instead of “marrow
juice” for cell kinetic analysis: Comparison of bone marrow biopsy
and aspiration material. Leuk Res 6:601, 1982
18. Holdrinet RSG, Van Egmond J, Wesse JMC, Haanen C: A
method for quantification of peripheral blood admixture in bone
marrow aspirates. Exp Hematol8:103, 1980
19. Clarkson B, Strife A, Fried J, Sakai Y, Ota K, Ohkita T,
Masuda R: Studies of cellular proliferation in human leukemia. IV.
Behaviour of leukemic cells in 3 adults with acute leukemia given
continuous infusions of ’H-thymidine for 8 to 10 days. Cancer
26:1237, 1970
20. Saunders EF, Mauer AM: Re-entry of non-dividing leukemic
cells into a proliferative phase in acute childhood leukemia. J Clin
Invest 48:1299,1969
21. Them1 H, Trepel F, Schick P, Kabolti W, Begemann H:
Kinetics of lymphocytes in chronic lymphocytic leukemia: Studies
using continuous ’H-thymidine infusion in two patients. Blood
42:623, 1973
22. Raza A, Imren S, Gezer S, Verkh L, Freeman JW: Expression of nucleolar antigen p145 in bone marrow cells of patients with
myeloid leukemias. Cancer Res 49:482, 1989
23. Raza A, Gezer S, Yasin Z, Mehdi I, Bhayana R, Black A:
Measurement of “growth fraction” (GF) from bone marrow (BM)
samples of patients with acute myeloid leukemia (AML) using a
monoclonal antibody (5C2) against a cycle specificprotein. Proc Am
Soc Hematol311a:100, 1988
24. Gezer S, Yasin Z, Imren S, Raza A: Changes observed in the
growth fraction, labeling index, duration of S-phase and total cell
cycle times of HL-60 cells as they undergo differentiation in response
to retinoic acid. Cancer Res 485989, 1988
25. Menger H, Yu-chen Y, Shu-rong C, Jin-ren C, Jia-Xiang L,
Lin Z, Long-jun G, Zhen-yi W: Use of all-trans retinoic acid in the
treatment of acute promyelocytic leukemia. Blood 72567, 1988
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1990 76: 2191-2197
Direct relationship between remission duration in acute myeloid
leukemia and cell cycle kinetics: a leukemia intergroup study
A Raza, HD Preisler, R Day, Z Yasin, M White, J Lykins, C Kukla, M Barcos, J Bennett and G
Browman
Updated information and services can be found at:
http://www.bloodjournal.org/content/76/11/2191.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.