impact of in vivo administration of interleukin 3 on ...€¦ · impact of in vivo administration...

Vol. 1, 295-303, March 1995 Clinical Cancer Research 295

Impact of in Vivo Administration of Interleukin 3 on Proliferation,

Differentiation, and Chemosensitivity of Acute Myeloid Leukemia’

Steven D. Gore,2 Li-Jun Weng, Richard J. Jones,

Kathleen Cowan, Milada Zilcha,

Steven Piantadosi, and Philip J. Burke

The Johns Hopkins Oncology Center, Oncology 2-109, Baltimore,

Maryland 21287-8963

ABSTRACT

Early clinical trials of growth factor augmentation of

induction chemotherapy for acute myeloid leukemia have

yielded variable results. To test the hypothesis that this

heterogeneity is a consequence of the pleiotropic effects of

growth factors on leukemic cell biology, we measured the

effects of in vivo interleukin 3 (IL-3) administration on

leukemic cell proliferation and drug sensitivity. Thirty-four

patients with acute myeloid leukemia with high-risk features

or advanced myelodysplasia received IL-3 as a continuous

infusion beginning 3 days prior to chemotherapy and con-

tinuing for the duration of intensive induction therapy. Bone

marrow cells were studied prior to and after 3 days of IL-3administration to assess changes in overall and leukemic

progenitor cell [leukemia colony-forming unit (CFU-L)]

proliferation, and progenitor cell sensitivity to 1-�3-D-

arabinofuranosylcytosine. The median fold increase in over-

all leukemic cell proliferation in response to IL-3, assessed asexpression of the nuclear antigen Ki67 in 28 patients, was

1.2. The median fold increase in percentage of cells in S

phase (assessed in 29 patients) was 1.3. Despite the increase

in overall cell proliferation in 70% of cases, CFU-L number

increased in only 4 of 20 patients successfully studied (me-

dian day 4:day 1 ratio of CFU-L number, 0.6). While thissuggests possible terminal differentiation of leukemic pro-

genitor cells, expression of CD34, HLA-DR, c-kit, CD15, andCD14 were not consistently affected by the cytokine. 1-I�-D-

Arabinofuranosylcytosine sensitivity of CFU-L increased

significantly in 30% of cases, decreased in 30%, and was

unchanged in 40%. Changes in overall cell proliferation(Ki67 expression) and CFU-L were independent predictors

of change in 1-f�-D-arabinofuranosylCytoSifle sensitivity; in-crease in percentage of cells in S phase in response to IL-3

was correlated with attainment of complete remission. While

these findings support the concept of cell cycle recruitment,

IL-3 has marked pleiotropic effects on proliferation, differ-entiation, and survival of leukemic progenitors which make

the clinical impact of in vivo cytokine administration for

individual patients difficult to predict.

INTRODUCTION

Because of the critical dependence of antileukemic chemo-

therapy on cycle-active cytotoxic drugs, recruitment of G11 leu-

kemic cells into active phases of the cell cycle may enhance the

cure rate in AML.3 In vitro models have shown enhanced killing

of leukemia cells by ara-C in the presence of myeloid growth

factors (1, 2), increased ara-C uptake and 1-�3-D-arabinofurano-

sylcytosine 5’-triphosphate retention following exposure to

GM-CSF (3), and increased selectivity of ara-C uptake in leu-

kemic cells when compared with normal hematopoietic cells (4,

5). Several Phase I and Phase II clinical trials of GM-CSF

augmentation of AML chemotherapy have demonstrated the

clinical feasibility of such an approach (6-9). Preliminary re-

sults of one randomized trial have suggested an improvement in

disease-free survival with GM-CSF (10), and a randomized

study of G-CSF augmentation of an unconventional induction

regimen (fludarabine plus ara-C) showed improved survival in

the G-CSF arm, although the effects of G-CSF were not con-

firmed in multivariate analysis (11). Conversely, data from the

M. D. Anderson Cancer Center (using historical controls) have

raised concerns about possible increased drug resistance in

patients receiving GM-CSF to augment induction chemotherapy

(8).

In 1992 we initiated a Phase I trial of IL-3 for cell cycle

recruitment in high risk and relapsed cases of AML and ad-

vanced MDS undergoing intensive ara-C-based remission in-

duction chemotherapy. Since IL-3 may affect growth and dif-

ferentiation of somewhat more primitive normal hematopoietic

cells than does GM-CSF, IL-3 may impact on the biology and

chemosensitivity of a broader phenotypic spectrum of AML

than does the latter cytokine. In fact, IL-3 stimulates the in vitro

proliferation of more cases of AML than does GM-CSF (12),

with a greater mean increase in proliferation index (13). In the

current trial, IL-3 was begun 3 days prior to chemotherapy and

continued for the duration of chemotherapy. The effects of IL-3

infusion on the proliferation, differentiation, and chemosensi-

tivity of leukemic blasts and clonogenic precursors were studied

Received 8/31/94; accepted 12/7/94.

I Supported by American Cancer Society Grant EDT-28. S. D. 0. is the

recipient of a Career Development Award from the American Cancer

Society. L. J. W. is the recipient of a Florence A. Carter Fellowship in

Leukemia Research from the American Medical Association Education

and Research Fund. R. J. J. is a Scholar of The Leukemia Society.

2 To whom requests for reprints should be addressed, at The JohnsHopkins Oncology Center, Oncology 2-109, 600 North Wolfe Street,

Baltimore, MD 21287-8963.

3 The abbreviations used are: AML, acute myeloid leukemia; AHD,antecedent hematological disorder; ara-C, 1 -�3-D-arabinofuranosylcy-

tosine; BrdUrd, bromodeoxyuridine; CFU-L, leukemia colony-forming

unit; CR, complete remission; FAB, French-American-British classifi-

cation of leukemia; GM-CSF, granulocyte-macrophage colony-stimu-

lating factor; G-CSF, granulocyte colony-stimulating factor; IL-3, inter-

leukin 3; LD99%, dose of drug which eliminates 99% of control colony

growth; MDS, myelodysplasia; PE, phycoerythrin; P1, propidium io-

dide; PR, partial response.

Research. on March 26, 2021. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

296 Impact of IL-3 on Acute Myeloid Leukemia

Tab le 1 Patient charact eristics an d impact of IL-3 on leukemic ccl 1 biology

IL-3. � receptor

. Entry (Percent Ki67 fold S-phase fold CFU-L fold ara-C sensitivity ClinicalPatient Age Sex FAB criteria Dose Karyotype Cells+) increase” increase” increase” fold increase” outcome

1 76 M Ml Age 2.5 -5, c” NDC �4T�1.5 2.7 4.86 NE

2 37 M M7 FAB, PRT 2.5 -Y, 7q+ ND 1.6 1.2 0.4 0.29 NR3 50 F M6 REL Rd 2.5 5q-,7q-, c ND 0.7 0.9 1.32” 1.5 NR4 43 F M6 FAB 2.5 -7 ND 1 1.9 0 NE CRS 66 M M7 FAll, PRT 5 Sq-, 7q-, c ND 1.4 1.3 7.32 0.5 PR6 46 M MO AHD 5 Sq-, c 30 3.1 8.7 0.34 2 CR7 67 M RAEB Secondary 5 -5, -7, c 14.6 1.2 0.4 0.29” 0.68 PR8 61 M RAEB Secondary 10 -5, c ND 1.8 0.9 No growth No growth NE9 66 F Ml Age 10 46XX ND 1 1.1 0.25 0.69 Tox

10 70 M MO Age 10 +8, c ND IL-3 D/C IL-3 D/C IL-3 D/C IL-3 D/C Tox11 20 F M7 REL Rd 10 5q- 5.7 1.1 1 No growth No growth NR12 41 M RAEB Secondary 10 -7, c 4.7 0.4 2.4 0.13 1.52 CR13 71 M Ml Age, AHD 10 -7, c 20 1.5 2 0.79 NE Tox14 70 F M4 Age 10 INV(9q) 0 1.2 3.7 l.1J4d 4.67 CR15 44 M M2REL Rd 10 5q-, -7,c 5 1.3 2.8 213 NE Tox16 69 F RAEB-t Age,

RAEB-t

20 -5 ND IL-3 D/C IL-3 D/C IL-3 D/C IL-3 D/C Tox

17 70 M RAEB-t PR 10 46 XY ND ND ND ND ND PR18 59 M MS Age 20 +8, +21 0 0.9 1.3 0.58 0.04 Tox19 55 F M4 REL Rd 10 46 XY ND Early Early Early Early NR

20 43 F21 32 F

M4 RELM6

RdFAB

1010

t(9;22)46 XX

0.244.4

chemotherapy11.2

chemotherapy1.51.3

chemotherapy145d

0.72”

chemotherapy10

0.05NR

NR22 30 F M4 AHD 10 -5, -7, c 1.2 1.4 2.4 1.56 0.32 CR23 48 F RAEB RAEB 10 46 XX 9.3 0.8 5.1 0.29 0.5 CR24 69 F Ml AFID 2.5 8 40 0.9 1.2 0.63” 1.16 NE25 54 F Ml AHD 10 46 XX 31.5 2.1 3.4 No growth No growth CR26 37 F M4 AHD 5 t(1;7) 03 1.7 0.6 0.16 1 NR27 34 F M4 REL Rd 5 46 XX 7 1.2 1.1 No growth No growth NR28 66 F29 62 M

RAEB-tM7

RAEBFAB

5

5

5q-, -7, c+13, ISO(17)

19

1.846

30.9

No growth

0.32”No growth

300CR

PR30 63 M M7 FAB 5 46 XY 1.7 1.1 1.2 No growth No growth NR31 64 M MO Age 5 46 XY <1 1.4 1.2 No growth No growth CR32 55 F M6 AFID 5 NE ND Fibrotic Fibrotic No growth No growth PR33 42 M M6 FAB 5 -5, c 1 1.1 1.7 No growth No growth CR34 74 F M5 Age 5 -7 ND 1.1 1.1 No growth No growth NR

a Fold increase, day 4 value: day 1 value.b c, other complex cytogenetic abnormalities present.C ND, not done; IL-3 D/C, IL-3 discontinued before study was completed; NE, not evaluable; NR, no response; Tox, treatment-related death;

PRT, primary refractory; Rel, relapse.d Difference between day 4 and day 1 value not statistically significant.

in bone marrow samples obtained prior to beginning the cyto-

kine infusion and again prior to the initiation of chemotherapy.

These studies were undertaken to test the hypothesis that the

variability of results of clinical trials of growth factor augmen-

tation are due to the pleiotropic effects of growth factors on

leukemic cell biology and to the biological heterogeneity of

AML.

MATERIALS AND METHODSPatients. All adult patients with MDS and AML referred

to The Johns Hopkins Oncology Center were assessed for eli-

gibility in this trial beginning in August 1992. Eligible patients

had AML in first relapse, AML which failed to go into remis-

sion after one course of infusional ara-C-based chemotherapy

(p rimary refractory disease), refractory anemia with excess

blasts, refractory anemia with excess blasts in transformation, or

newly diagnosed AML with one of the following high-risk

criteria: history of AHD, FAB subtype M6 or M7, secondary

AML, or age �55 years. Patients were excluded for any of the

following criteria: Karnofsky performance status <60%; total

WBC >100,000/pA, serum creatinine �2.0 mg/dL; total biliru-

bin �2.5 mg/dl; aspartate aminotransferase, alanine aminotrans-

ferase, or alkaline phosphatase >5 times normal range; chemo-

therapy or growth factor administration within 4 weeks prior to

study; ongoing disseminated intravascular coagulation; clini-

cally evident leukostasis; and positive serum f3-human chorionic

gonadotropin. All patients gave written informed consent as

approved by the Institutional Review Board under Department

of Health and Human Services guidelines.

Clinical characteristics of the 34 treated patients are listed

in Table 1. The median age was 55 years, and 20 patients had

karyotypic abnormalities associated with poor prognosis (14).

While several patients had more than one high-risk feature

making them eligible for the protocol, the primary (i.e., most



Clinical Cancer Research 297

potentially clinically significant) high-risk features were as fol-

lows: relapse (n = 7), primary refractory leukemia (n = 3), FAB

M6 or M7 (n = 4), history of AHD (n = 8), advanced MDS

(n = 5), and age (n 7).

Treatment. Patients were treated with IL-3 at one of the

following doses (�i.gIkg/day): 2.5, 5.0, 10.0, 20.0 (Table 1). IL-3

was produced by Sandoz (East Hanover, NJ) and supplied

through the Cancer Therapy Evaluation Program of the National

Cancer Institute. IL-3 was given as an iv. continuous infusion

from days 1 to 13. Chemotherapy was given during the IL-3

infusion as follows: ara-C, 0.667 g/m2/day on days 4-6 as an

iv. continuous infusion; daunorubicin, 45 mg/m2/day on days

4-6 as an i.v. injection; etoposide, 400 mg/m2/day on days

1 1-13 as a 4-6-h iv. infusion. If the initial WBC count was

>50,000/pA, chemotherapy was begun concurrently with the

IL-3 infusion on day 1. If the WBC count rose to >75,000/pA

before day 4, chemotherapy was begun early. Bone marrow

aspirates were obtained before beginning the IL-3 infusion (day

1) and during the IL-3 infusion before beginning chemotherapy

(day 4). If chemotherapy was initiated early, a second bone

marrow was obtained at that time when possible. Supportive

care was provided as previously described (15). Clinical details

and outcome of the trial will be presented in a subsequent

publication.

Cells. Mononuclear cells from heparinized samples of

bone marrow aspirates were obtained by density centrifugation

(specific gravity, <1.077 g/dl, Ficoll-Hypaque; Pharmacia, NJ)

and used on the same day for all assays. Cytocentrifuge prepa-

rations of the mononuclear cells were stained with Wright’s

stain and differential blood counts performed manually. The

myeloid leukemia cell line KG1a was maintained in RPMI 1640

medium (Sigma, St. Louis, MO) and maintained at 37#{176}Cunder

5% CO2 as a control cell line for proliferation assays.

Proliferation Assays. The Ki67 antigen, a nuclear pro-

tein of unknown function, was originally described in phytohe-

magglutinin-stimulated lymphocytes, and was shown to be cx-

pressed as cells entered G1 from G0 with continued expression

throughout the cell cycle (16, 17). Ki67 expression was mea-

sured in cells by a flow cytometric assay as described previously

(18, 19). To exclude contaminating T cells from the analysis,

cells were first stained with PE-labeled CD5 antibody (leul-PE;

Becton Dickinson, Mountain View, CA) or PE-labeled isotype-

matched control (Becton Dickinson) before fixation and perme-

abilization (18, 19). Anti-Ki67 and control mouse monoclonal

IgGi antibodies were purchased from Dakopatts (Carpinteria,

CA); affinity-purified FITC-labeled goat anti-mouse IgGi see-

ondary antibody was from Southern Biotechnology Associates

(Birmingham, AL). Logarithmically growing KG1a myeloid

leukemia cells were used as positive controls in all experiments.

A FACScan flow cytometer (Becton Dickinson) was used to

acquire 10,000 events in list mode using FACScan research

software following compensation of the PE and FITC channels

using Calibrite beads (Becton Dickinson) according to the man-

ufacturer’s instructions. Significant Ki67� signals were deter-

mined based on staining of identical cells with the control

primary antibody. CDS+ signals were excluded from the anal-

ysis. We have previously shown that Ki67 expression in marrow

mononuclear cells from patients with AML measured in this

assay correlates well with other markers of proliferation (19).

S-phase cells were determined using flow cytometry in a

modification of Dolbeare’s technique (20) following a 1-h pulse

with BrdUrd (10 ELM, 37#{176}C,Sigma). Cells were labeled with

FITC-conjugated anti-BrdUrd and the DNA-binding compound

P1 as described previously (19). Identically treated AML sam-

ples pulsed with medium instead of BrdUrd served as negative

controls, and KG1a cells pulsed with BrdUrd were positive

controls for the BrdUrd assay. Acquisition and analysis of the

BrdUrd/PI-stained samples was performed using Cell-FIT soft-

ware (Becton Dickinson). Analysis was performed using an

electronic gate based on P1 fluorescence pulse area and pulse

width measurements to exclude cell doublets and aggregates

according to the manufacturer’s instructions. Identical cells

pulsed with medium instead of BrdUrd were used to determine

the cutoff for BrdUrd expression, and all cells with significant

BrdUrd signal were considered to be in S phase.

Clonogenic Assay and ara-C Sensitivity. CFU-L were

quantified in methyl cellulose using a modification of McCul-

loch’s technique as described previously (21). Day 7 phytohe-

maglutinin-stimulated lymphocyte-conditioned medium was

used as a source of growth factors, and CFU-L were quantified

on day 5. CFU-L numbers were tested for significant impact of

IL-3 infusion using two-tailed Student’s t test of the means of

colony counts from triplicate or quadruplicate cultures plated on

day 4 compared to CFU-L numbers on day 1 of cytokine

infusion (P < 0.05). To determine the sensitivity of CFU-L to

ara-C, the drug was incorporated into the methyl cellulose plates

at a doses ranging from 0.5 to 5 p.M. Each dose was studied in

triplicate or quadruplicate. Semi-log dose-response curves of the

percentage of surviving CFU-L versus ara-C dose were plotted

and ara-C sensitivity was described as the slope of the regres-

sion line of the dose-response curve using least-squares analysis

of the log-transformed data (21-23). Significant modulation of

ara-C sensitivity by IL-3 was determined by comparing the 95%

confidence intervals of the slopes on day 1 and day 4; samples

were considered different if there was no overlap of the 95%

confidence intervals.

Immunophenotyping and Detection of Growth Factor

Receptor Expression. Surface antigens were detected follow-

ing direct or indirect immunofluorescence staining and fixation

in 1% formaldehyde as described previously (24). c-kit and IL-3

receptor expression were measured following binding of bio-

tinylated (Steel Factor) or PE-labeled (IL-3) growth factor (R

and D Systems, Minneapolis, MN) as described (13). Strepta-

vidin-PE was purchased from R and D Systems. Control cells

were incubated with streptavidin-PE only. Immunofluorescence

was measured on a FACScan flow cytometer using FACScan

Research Software; 10,000 events were acquired in list mode for

each sample. Percentage of positive cells was determined using

isotype-matched fluorochrome-labeled irrelevant mAbs or

streptavidin-PE (for biotinylated samples) as controls. Cases

were considered positive for a particular cell surface antigen if

�20% of cells demonstrated significant specific fluorescence;

growth factor receptors were considered positive if �1O% of

cells expressed significant fluorescence.

Additional Antibodies. FITC-labeled anti-HLA-DR

(CR3/43), CD1S (C3D-1), PE-labeled CD14 (TUK4), and iso-

type-matched control antibodies were from Dakopatts. PE-la-



Aa

0

I,

a0

F...(0

C

aCa

0

I-

a0

a0

aaa

a.

C

ag’a

0

Fig. I Impact of IL-3 administration on proliferation of leukemic

cells. Proliferation was assessed as the percentage of cells in cycle (Ki67

assay, A) and percentage of cells in the S phase (B). Data represent the

ratio of the day 4:day 1 value for each patient. Data are grouped

according to the direction of change: increased, no change, or decreased.

Change was considered significant if the value was increased or de-creased by at least 10% (i.e., fold increase of 1.1 or 0.9). No attempt has

been made to represent individual patients with the same symbol con-

sistently between figures.

U�U

+

U

Incr.a..d Unchang.d Dcrsas�

Direction of Change


beled CD34 antibody (HPCA-2) was purchased from Becton

Dickinson.

Clinical Response Assessment. Response was assessed

by standard criteria (25). CR required normalization of blood

counts with <5% blasts in the bone marrow and none in the

peripheral blood. PR was scored if the bone marrow was cleared

of blasts but evidence of myelodysplasia remained, either be-

cause of peripheral blood counts which stabilized in a nontrans-

fusion-dependent range but did not meet the criteria for CR, or

because of ongoing dysmorphic features in the bone marrow. In

addition, PR was reported if previous bone marrow fibrosis

persisted, despite the achievement of normal blood counts and

differential WBC count.

Statistical Methods. The major statistical outcomes of

this study were changes between day 1 and day 4 measurements

of hematological variables, proliferative parameters, CFU-L

number, and ara-C sensitivity. Changes in outcomes were cal-

culated in two ways: as a ratio of day 4 value divided by day 1

value and as a logarithm of the ratio (i.e., difference on a log

scale). Because of skewness in the distributions, we emphasize

results from calculating the differences on a log scale. A two-

tailed paired Student’s t test of the log-transformed values was

used to test differences between means; means are reported ±

SE. Differences between medians were tested using the Wil-

coxon signed rank test (paired samples) or the Wilcoxon rank

sum test (unpaired samples). Pairwise associations between

outcome variables were summarized using correlation coeffi-

cients and multivariate associations were modeled using linear

regression. Variables which were found to correlate well in

pairwise analyses (r > 0.4) were entered into multiple regres-

sion models. Nonsignificant predictors were then removed from

the model one at a time with reestimation of regression slopes

and significance levels at each step (step-down procedure).

RESULTS

Effect of IL-3 on Leukemic Cell Proliferation. Bone

marrow samples before and after IL-3 treatment were obtained

from 29 of 34 patients. Day 4 marrows were not obtained from

four patients for the following reasons: discontinuation of IL-3

(n 2), urgent chemotherapy initiated early because of a

rapidly rising WBC count (n = 1), and fibrotic bone marrow

with insufficient cells obtained on day 1 for adequate study (n

1). No patient had chemotherapy started simultaneously with

IL-3. Ki67 expression was highly variable on both days 1 and 4,

ranging from 1 to 84% on day 1 and 1 1 to 92% on day 4

(median, 40 and 45, respectively; P < 0.01, signed rank test).

The ratios of day 4:day 1 for individual patients are shown in

Fig. IA. In this and subsequent Figs. 2-4, the patients are

grouped according to whether the percentage of Ki67� cells

significantly increased, decreased, or was not significantly

changed on day 4 compared with day 1. The mean ratio of day

4:day 1 Ki67 expression was 1.4 ± 1.1 (P = 0.04; median ratio,

1.2). The change in Ki67 expression did not correlate with the

dose of IL-3 administered. Ki67 expression was also not corre-

lated with FAB classification (MO-MS versus M6 or M7),

history of AHD, or whether patients were relapsed.

IL-3 administration impacted on the S-phase fraction sim-

ilarly to the Ki67� subset. Percentage of cells in S phase ranged

100

0

10

A

1 � 0.+#{149}+#{149}

U

0.1 ‘ � I

Mcrus.d Unchang.d Decr#{149}aud

Direction of Change

Bic A

0

V �0

�vV0

�

from 1 to 23 on day 1 and 1.7 to 20 on day 4 (median, 5.8 and

8.7, respectively; P < 0.01, signed rank test). Day 4:day 1 ratios

of the percentage of S-phase cells for individual patients are

shown in Fig. lB. The mean day 4:day 1 ratio of the S-phase

percentage was 1.6 ± 1.1 (P < 0.008; median ratio, 1.3). The

change in the S-phase percentage was not correlated with the

dose of IL-3 administered or the clinical parameters noted above

for Ki67.

Changes in Clonogenic Leukemic Cell Proliferation.

Because the proliferative status of the bulk of bone marrow

leukemia cells may not reflect changes in proliferation among

clonogenic leukemia cells, changes in CFU-L number were

measured before and after 3 days of IL-3 infusion. CFU-L were

successfully cultured from the bone marrows of 21 patients

(70%). The cloning efficiency of patient samples varied widely,



a0

a0

ii

II-0

C

aC

a0

+

10

0+

:�:�

VV0.1

0.01 00

0.001 ‘ I I

1000

100

10

0.1

0.01

>1a0

a0

aCa

(1�

a

C

aDCa

0

0+

‘�

00

$

0

I

Incr.assd Unckng.d D#{149}cr.aud


Incrsased Unchsngd Dcr#{149}assd

Direction of Change

Fig. 2 Impact of IL-3 administration on CFU-L number. The mean

CFU-L number from triplicate or quadruplicate cultures for each patientplated on day 4 was divided by the mean CFU-L number from cultures

plated on day 1. Data are grouped according to the direction of change:

increased, no change, or decreased. Change in CFU-L was considered

significant if the mean CFU-L number on day 4 was significantlydifferent from that on day 1 using two-tailed Student’s t test (P � 0.05).

Thus there is some apparent overlap in the magnitude of change between

patients in the ‘ ‘no change’ ‘ group (whose mean values on the 2 dayswere not significantly different) and in each of the other groups (whose

mean values were significantly different).

ranging from 0 to 1600 colonies/105 cells on day 1 (median,

130) and 0 to 1150 on day 4 (median, 58). The ratios of day

4:day 1 CFU-L number are plotted in Fig. 2. Significance in

change was determined by comparing the means of triplicate or

quadruplicate cultures on days 1 and 4 using a two-sided Stu-

dent’s t test (see ‘ ‘Materials and Methods’ ‘). CFU-L number

increased significantly in 4 patients (19%), decreased signifi-

cantly in 12 (57%), and was unchanged in 5 (23%). The median

change in colony formation was a decrease of 40% on day 4.

Differential cell counts of Wright-stained cytocentrifuge

preparations of BMMC from days I and 4 showed that change

in percentage of blasts ranged from -60% to +49%, with a

median change of +3%. The mean ratio of day 4:day 1 blast

percentage was 0.6 ± 1.6 (P = not significant). Change in

CFU-L number was not correlated with change in blast percent-

age when tested as continuous variables or when direction of

change was compared through x2 analysis. The change in

CFU-L number was also not correlated with change in Ki67

expression, IL-3 dose administered, FAB classification, history

of AHD, or relapsed AML.

Effect of IL-3 Infusion on Chemosensitivity of CFU-L.

The impact of the cytokine infusion on ara-C sensitivity was

assessed in 18 patients (62%). Three patients in whom colony

formation could be assessed formed too few colonies on either

day 1 or day 4 to reliably quantify ara-C sensitivity on that day.

The day 4:day 1 ratios of ara-C sensitivity are plotted in Fig. 3

according to the significance and direction of change of sensi-

tivity. Significance of change in ara-C sensitivity was deter-

mined by comparing the 95% confidence intervals of the slopes

Direction of Change

Fig. 3 Impact of IL-3 administration on ara-C sensitivity of CFU-L.

ara-C sensitivity is reported as the slope of the log-linear dose response

curve of ara-C. A slope of 1 or greater is considered clinically relevant

ara-C sensitivity (corresponding to a LD99% of 2 p�; Ref. 26). Data

represent the ratio of day 4:day 1 slopes calculated from the regression

lines. Data are grouped according to the direction of change. Change in

ara-C sensitivity was considered significant if the 95% confidence

intervals of the slopes on day 4 and day I did not overlap. As in Fig. 2,

there is some overlap in the magnitude of change between patients in the

‘ ‘no change’ ‘ group (whose 95% confidence intervals of the dose-

response slopes on day 1 and day 4 overlapped) and in the other two

groups (whose 95% confidence intervals on the 2 days did not overlap).

of the ara-C dose-response curves on days 1 and 4 (see ‘ ‘ Ma-

terials and Methods’ ‘). Using a slope of the ara-C dose-response

curve of 1 as a cutoff for clinically relevant ara-C sensitivity

(corresponding to a LD99% of 2 pM; Ref. 26), 9 of 18 samples

were sensitive to ara-C on day 1; 9 were innately resistant. The

median ara-C sensitivity on day 1 was 0.2 (range, 0.01-24).

After 3 days of IL-3 infusion, the median ara-C sensitivity was

0.7 (range, 0.01-26). Five patients (patients 1, 14, 15, 20, and

28) had significant increases and five patients (patients 9, 18, 21,

22, 23) had significant decreases in ara-C sensitivity, with eight

patients exhibiting no significant change. Two ‘ ‘ resistant’ ‘ cases

were brought into the ‘ ‘sensitive’ ‘ range (slope � 1), while two

sensitive cases demonstrated in vitro resistance to ara-C on day

4. Among the patients who demonstrated significant increases in

ara-C sensitivity, the median ratio of day 4:day I ara-C sensi-

tivity was 10 (range, 5-300); among those patients with signif-

icant decrements in ara-C sensitivity, the median ratio was 0.3

(range 0.04-0.7). No correlation was found between change in

ara-C sensitivity and IL-3 dose.

Because of the marked heterogeneity of change in CFU-L

ara-C sensitivity following IL-3 infusion, possible correlations

between the log difference between day 4 and day 1 ara-C

sensitivity and the log difference in Ki67, S phase, and CFU-L

expression were tested. The change in ara-C sensitivity corre-

lated with both the change in Ki67 (r = 0.55; P = 0.02) and

change in CFU-L number (r 0.4; P = 0.09) (Fig. 4). Multi-

variate analysis using these two variables to predict the change



Aa0

a0

aCa(0

a

aCaU

0

a0

� 5

a0

aCa(0 -�U

Log Change K167 (Day 4 - Day 1)

UU


-5

UU

UUU

-7 . . . . , .

-3 -2 -1 0 1 2 3 4 5 6

! Log Change CFU-L (Day 4 - Day 1)

Fig. 4 Correlation between proliferation and ara-C sensitivity. Thedifferences between the day 4 and day 1 log-transformed data for Ki67expression, CFU-L number, and ara-C sensitivity (corresponding to aratio of day 4:day 1 parameters) were calculated and plotted. The lines

represent results of linear regression. A, correlation between change in

Ki67 expression and change in ara-C sensitivity (slope, 1.3; P 0.02).

B, relationship between changes in CFU-L number and ara-C sensitivity

(slope, 0.5; P = 0.09). In multivariate analysis, the P values for these

two relationships were 0.01 and 0.05, respectively.

in ara-C sensitivity showed that both change in Ki67 and change

in CFU-L were independent predictors of change in sensitivity.

Change in ara-C sensitivity varied with change in Ki67 with a

slope of 1.3 (P = 0.01); the relationship with change in CFU-L

had a slope of 0.5 (P = 0.05). Although FAB classification or

relapse status had no detectable impact on the change in ara-C

sensitivity, de novo patients had a greater median day 4:day 1

ara-C sensitivity ratio than did patients with either a history of

AHD or with MDS (median day 4:day 1 ratios, 4.75 versus 0.6;

P = 0.05, Wilcoxon rank sum test).

Immunophenotypic Evidence of Differentiation. Be-

cause the number of clonogenic cells decreased on day 4 in

many samples, immunophenotypic evidence of differentiation

was sought in the more recently enrolled patients. In these

patients, bone marrows from days 1 and 4 were studied for

expression of the progenitor cell antigens c-kit, CD34, and

HLA-DR, and the markers of monocytic and granulocytic dif-

ferentiation CD14 and CD15 (Table 2). In normal myeloid

differentiation, CD34, c-kit, and HLA-DR cease to be cx-

pressed, while levels of CD14 and/or CD15 increase, depending

on the cell type. CD34 expression decreased significantly in

only 1 of 12 patients studied. HLA-DR expression decreased in

no patients; however, c-kit was down-regulated in 5 patients

(median decrease, 40%). Four of five patients with significant

decrements in c-kit expression had significant decreases in

CFU-L growth on day 4 (the fifth patient demonstrated no

CFU-L growth on either day). In contrast, two patients showed

increased expression of CD34, HLA-DR, and c-kit, and three

additional patients increased c-kit expression alone. No patient

changed CD1S expression by more than 10% between days 1

and 4. Of the 2 of 12 patients who expressed CD14 on day 1, 1

increased expression and 1 decreased expression of this antigen

on day 4.

Predictors of Biological and Clinical Response. Twen-

ty-one patients (69%) had IL-3 receptor expression quantified

using a flow cytometric assay utilizing binding of PE-labeled

ligand. Twelve of these patients had detectable IL-3 receptors

(range, 4-40% cells positive; median, 9%). Of these patients,

21 were evaluable for changes in Ki67 expression, 15 for

changes in CFU-L number, and 13 for changes in ara-C sensi-

tivity. Changes in these three parameters did not correlate with

the percentage of cells which expressed measurable IL-3 recep-

tors on day 1.

Twenty-five patients could be evaluated for clinical re-

sponse to cytokine-augmented chemotherapy (Table 1). Of the

nine nonevaluable patients, three patients had IL-3 discontinued

due to toxicity or deteriorating performance status. The other six

died during remission induction. Of the 25 evaluable patients,

12 achieved CR, S were considered PR, and 8 were nonrespon-

sive. Fourteen evaluable cases also had ara-C sensitivity suc-

cessfully assessed on day 1. Three of five patients with day 1

ara-C sensitivity � 1 (corresponding to a ‘ ‘drug-sensitive”

LD99% of >2 p.M; Ref. 26) achieved CR; three of nine patients

with day 1 ara-C sensitivity < 1 achieved CR. Twelve of the

evaluable cases also had day 4 ara-C sensitivity successfully

assessed. Only one of six patients whose ara-C sensitivity on

day 4 was <1 achieved a CR. In contrast, four of six patients

whose day 4 ara-C sensitivity was �1 achieved CR. Of these

four patients, three had ara-C sensitivity in the sensitive range

on day 1 as well. Of the two patients with day 4 ara-C sensitivity

�1 who did not achieve CR, 1 had ara-C sensitivity of 1 on day

1; the other patient was resistant on day 1. The trend between

remission achievement and day 4 ara-C sensitivity did not

achieve statistical significance (P = 0.12, Fisher’s exact test).

Of the other biological parameters studied, only the change

in percentage of cells in S phase predicted for remission

achievement with patients whose day 4:day 1 ratio of S-phase

cells was �1.16 more likely to achieve CR (P < 0.004, Fisher’s

exact test).




Table 2 Immunophenotypic evidence of differentiation”

Antigen

No/No . studied % Cells� median (range)

No. increased No. decreasedDay 1 Day 4 Day 1 Day 4

CD34HLA-DRc-kit

CD1SCD14

7/136/12

13/18

4/122/12

5/127/1 1

11/17

6/111/11

40 (22-82) 75 (32-93)

80 (57-96) 88 (26-95)50 (15-94) 54 (24-94)

36 (20-80) 30 (21-89)(23,39) (53)

2

2

5

01

1

0

5

01

a Bone marrow mononuclear cells from days 1 and 4 of cytokine administration were studied for antigen expression by two-color flow

cytometry. Cases were considered positive if �2O% of cells expressed significant fluorescence (� 10% in the case of c-kit). Significant increaseor decrease in antigen expression was defined as a change in cellular expression of at least 10%.

DISCUSSION

The failure of current aggressive chemotherapy regimens

to cure the majority of patients with AML, particularly those

cases evolving from myelodysplasia and those with unfavorable

FAB classifications or cytogenetic abnormalities (14) requires

the development of new therapeutic strategies. In the absence of

promising new chemotherapeutic agents, appropriate biomodu-

lation of AML chemotherapy could positively impact on out-

come in these diseases. The use of myeloid growth factors to

recruit leukemic progenitor cells into cycle, sensitizing the cells

to the cytotoxic effects of cycle-specific agents such as ara-C

and, to a lesser extent, the topoisomerase II inhibitors, has

grown out of a long history of preclinical and clinical studies

suggesting the utility of this approach (1-4, 15, 27-29). How-

ever, the results of the early trials of growth factor augmentation

of induction chemotherapy for AML have been variable.

The heterogeneity of these trial outcomes may reflect the

pleiotropic biological effects of growth factors, as well as the

heterogeneity of the diseases which are lumped together as

AML. The effects of hematopoietic growth factors on the biol-

ogy of normal and leukemic cells can be categorized into effects

on proliferation, survival, and differentiation. Increased cycling

of leukemia progenitor cells in response to growth factors would

potentially increase their sensitivity to cytotoxic agents. How-

ever, recent evidence demonstrates that hematopoietic growth

factors inhibit apoptosis (30), the primary mechanisms by which

cells die in response to cytotoxic agents. Because inhibited

apoptosis resulting from multiple different processes can pro-

duce drug resistance (26, 31-34), growth factor administration

could actually decrease drug sensitivity. Differentiation of leu-

kemic progenitor cells could be a beneficial affect of growth

factor administration if this was associated with a loss in self-

renewal capacity, but might have variable effects on drug sen-

sitivity. In an in vitro model, myeloid growth factors shifted the

growth of AML colony-forming cells from ‘ ‘self-renewing’ ‘ to

more differentiated colony growth, associated with a decrease in

ara-C sensitivity (35). Thus, the clinical impact of growth factor

administration in conjunction with chemotherapy for AML for

an individual patient may be a complex function of the relative

effects of the cytokine on proliferation, survival, and differen-

tiation in that case. These effects in turn reflect the interactions

of numerous intracellular pathways normally or aberrantly slim-

ulated or inhibited by that particular cytokine in a particular case

of AML.

In the current Phase I study of IL-3 in poor-risk AML, the

pleiotropic effects of this cytokine were clearly demonstrated.

While the dose of IL-3 administered to evaluable patients

ranged from 2.5 to 10 p.g,’kg/day in this Phase I study, biological

effects of the cytokine were seen at all doses, with no trend

toward a dose-response effect. Hence it is likely that the cyto-

kine has significant activity at all dose levels tested, and the

variability in outcomes reflects the heterogeneity of the response

of leukemic cells to the cytokine. The biological changes mea-

sured could not be predicted by percentage of cells with signif-

icant IL-3 receptor expression detected by the flow cytometric

assay. The biological response to IL-3 and other myeloid growth

factors has been demonstrated in vitro to correlate poorly with

receptor expression measured by a variety of assays; significant

biological responses have been seen in the absence of detectable

receptors. It is likely that small numbers of IL-3 receptor are

necessary to effect biological responses (13, 36, 37).

IL-3 moderately stimulated proliferation of leukemic cells

in the majority of cases at all doses tested (median increase,

20-30%); however, the numbers of clonogenic leukemic cells

decreased during the 3-day cytokine infusion in most cases in

which colonies could be successfully grown. The decrease in

CFU-L number despite increased proliferation in the majority

leukemic cell population suggests that proliferation is associated

with ongoing differentiation of the progenitor cells. Evidence of

differentiation of the majority of the leukemia cells was modest:

no net effect of the cytokine infusion was seen on the median

blast percentage (assessed morphologically); CD34 expression

decreased in only 1 of 13 patients studied, with no patient

demonstrating decreased HLA-DR expression. CD14 expres-

sion increased in one patient whose leukemia expressed this

antigen; no significant impact was seen on CD15 expression.

Down-regulation of c-kit expression in 5 of 15 patients studied

may represent a differentiation event; however, the biological

impact of expression of this receptor on AML cells has not been

extensively studied.

The in vitro assay for ara-C sensitivity used in this study

has previously been shown to correlate with remission attain-

ment (26); in the small sample size in which clinical response

could be compared with in vitro ara-C sensitivity in the present

study, a trend was seen toward remission attainment in patients

whose ara-C sensitivity was �1 on day 4 of therapy. The

correlation with remission attainment must be viewed in the

context of the fact that all patients were treated with two active

agents besides ara-C, the topoisomerase Il-active drugs dauno-

rubicin and etoposide. Although topoisomerase II expression




varies during the cell cycle and increases after in vivo treatment

with GM-CSF (38), we did not measure the impact of IL-3 on

the CFU-L sensitivity to these drugs. Instead, we restricted our

in vitro investigation to the effects of the cytokine on ara-C

sensitivity, because ara-C forms the foundation of most first-line

treatments for AML, and as the most cell cycle dependent of the

three agents would be most likely to be influenced by changes

in cell proliferation. IL-3 significantly altered clonogenic ara-C

sensitivity in almost two-thirds of patients from whom colonies

could be cultured; however, ara-C sensitivity was increased and

decreased in equal numbers of cases.

The mechanism of increased ara-C resistance in a signifi-

cant subset of patients is uncertain. The role of IL-3 as a survival

(anti-apoptotic) factor may cause it to function as a chemopro-

tectant in these patients. Hematopoietic growth factors directly

protected two mouse myeloid leukemia cell lines from chemo-

therapy-induced apoptosis (including ara-C; Ref. 39). Similar

protective effects of IL-3 on killing of human factor-dependent

AML cell lines and individual samples of primary AML by

superpharmacological doses of ara-C (120 p.M) have been dem-

onstrated (40). Besides protection against apoptosis, recent data

have suggested that driving cell cycling might paradoxically

decrease ara-C sensitivity by increasing deoxynucleotide pools

and thereby inhibiting deoxycytidine kinase, the rate-limiting

enzyme for 1 -�3-D-arabinofuranosylcytosine p -triphosphate

generation (41). However, the positive correlations demon-

strated in the present study between changes in overall (Ki67) or

clonogenic (CFU-L) proliferation and ara-C sensitivity make

this latter explanation less likely.

The independent positive correlations between the changes

in proliferation (Ki67 expression) and CFU-L number with

change in ara-C sensitivity support the recruitment hypothesis.

This is further supported by the apparent increased likelihood of

achieving CR in patients in whom IL-3 increased the percentage

of cells in the S phase by at least 16% (of the day 1 value). The

small size of the data set precludes establishment of change in

the S phase as an independent predictor. A major reason for the

failure of IL-3 to impact positively on ara-C sensitivity in many

of the cases may relate to its dual effects on proliferation and

differentiation. While the majority of cases showed increased

proliferation with IL-3, favoring increased ara-C sensitivity, the

number of clonogenic cells decreased in many cases, potentially

due to differentiation of clonogenic precursors, and favoring

decreased ara-C sensitivity (Fig. 4B). The clinical impact of this

putative differentiation effect apart from drug sensitivity is

unknown. Enrollment in the present study was limited to pa-

tients with high-risk leukemia; 50% of the patients studied were

innately ara-C resistant before beginning therapy (slope of dose-

response curve <1; LD99% >2 p.t�t; Ref. 26). Purely “kinetic

resistance’ ‘ may play a more important role in good-risk de

novo leukemias. In the present study, patients with a history of

AHD or who were treated with a diagnosis of MDS were less

likely to have increased ara-C sensitivity following IL-3 than

were patients with de novo leukemia. Nonetheless, the impact of

cytokine administration on differentiation and survival of leu-

kemic progenitor cells will need to be considered carefully in de

novo leukemia as well.

The biological heterogeneity of AML makes uniform ap-

proaches to all patients unlikely to succeed. Progress in AML

will more likely be made as therapy can be individualized based

on the biology of the particular case of leukemia. Because of the

heterogeneity of biological response to IL-3, future trials of

cytokine augmentation of chemotherapy effect may be best

planned following in vitro investigation of the cytokine effect on

chemosensitivity. As the intracellular pathways which regulate

proliferation, differentiation, and survival in hematopoietic cells

become further elucidated, leukemic cells may eventually be

screened at diagnosis to determine which patients are likely to

benefit most from cytokines administered concurrent with or

following chemotherapy, and those patients for whom cytokine

administration may induce drug resistance and is best avoided.

ACKNOWLEDGMENTS

We gratefully acknowledge the expert technical assistance of Mary

C. Johnson, the outstanding support of the technologists in The JohnsHopkins Oncology Center Hematology and Flow Cytometry Laboratory

under the direction of Peter Nicholls, and the administrative support of

Lisa Allen Minnick.

REFERENCES

1 . Cannistra, S. A., Groshek, P., and Griffin, J. D. Granulocyte macro-

phage colony stimulating factor enhances the cytotoxic effect of cy-

tosine arabinoside in acute myeloblastic leukemia and in the myeloid

blast crisis of chronic myeloid leukemia. Leukemia, 3: 328-334, 1989.

2. Lista, P., Procu, P., Avanzi, G. C., and Pegoraro, L. Interleukin-3enhances the cytotoxic activity of 1-beta-D-arabinofuranosylcytosine

(ara-C) on acute myeloblastic leukaemia (AML) cells. Br. J. Haematol.,69: 121-123, 1988.

3. Karp, J. E., Burke, P. J., and Donehower, R. C. Effects of rhGM-CSFon intracellular ara-C pharmacology in vitro in acute myelocytic leuke-mia: comparability with drug-induced humoral stimulatory activity.

Leukemia, 4: 553-556, 1990.

4. Bhalla, K., Holladay, C., Arlin, Z., Grant, S., Ibrado, A. M., and

Jasiok, M. Treatment with interleukin-3 plus granulocyte-macrophagecolony-stimulating factors improves the selectivity of ara-C in vitro

against acute myeloid leukemia blasts. Blood, 78: 2674-2679, 1991.

5. Bhalla, K., Birkhofer, M., Arlin, Z., and et al. Differential effect of

interleukin 3 on the metabolism of high-dose cytosine arabinoside innormal versus leukemic bone marrow cells. Exp. Hematol., 19: 669-

673, 1991.

6. Bettelheim, P., Valent, P., Andreeff, M., Tafuri, A., Haimi, J.,Gorischek, C., Muhm, M., Sillaber, C., Haas, 0., Vieder, L., Maurer, D.,Schulz, G., Speiser, W., Geissler, K., Kier, P., Hinterberger, W., andLechner, K. Recombinant human granulocyte-macrophage colony-stim-ulating factor in combination with standard induction chemotherapy inde novo acute myeloid leukemia. Blood, 77: 700-711, 1991.

7. Burke, P. J., Wendel, K. A., Nicholls, P. D., Hall, K. L., Morrall, L.,and Bonnem, E. M. A phase I trial of granulocyte-macrophage stimu-lating factor (GM-CSF) and humoral stimulating activity (HSA) as

biomodulators of timed sequential therapy (TST) of leukemia (AML).Blood, 76 (Suppl. 1): 1023a, 1990.

8. Estey, E., Thall, P. F., Kantarjian, H., O’Brien, S., Koller, C. A.,Beran, M., Gutterman, J., Deisseroth, A., and Keating, M. Treatment ofnewly diagnosed acute myelogenous leukemia with granulocyte-mac-rophage colony-stimulating factor (GM-CSF) before and during contin-

uous-infusion high-dose ara-C + daunorubicin: comparison to patients

treated without GM-CSF. Blood, 79: 2246-2255, 1992.

9. Archimbaud, E., Fenaux, P., Reiffers, J., Cordonnier, C., Leblond,V., Travade, P., Troussard, X., Tilly, H., Auzanneau, G., Marie, J-P.,Ffrench, M., and Berger, E. Granulocyte-macrophage colony-stimulat-ing factor in association to timed-sequential chemotherapy with mitox-

antrone, etoposide, and cytarabine for refractory acute myelogenousleukemia. Leukemia, 7: 372-377, 1993.




10. Buchner, 1., Hiddemann, W., Wormann, B., and et al. Recombinanthuman GM-CSF priming and long-term administration together withmultiple course chemotherapy (CT) vs CT alone in newly diagnosedacute myeloid leukemia (AML). Blood, 78 (Suppl 1): 163a, 1991.

11. Estey, E., Thall, P., Andreeff, M., Beran, M., Kantarjian, H.,O’Brien, S., Escudier, S., Robertson, L. E., Koller, C., Komblau, S.,Pierce, S., Freireich, E. J., Deisseroth, A., and Keating, M. Use ofgranulocyte colony-stimulating factor before, during, and after fludara-bine plus cytarabine induction therapy of newly diagnosed acute my-elogenous leukemia or myelodysplastic syndromes: comparison withfludarabine plus cytarabine without granulocyte colony-stimulating fac-tor. J. Clin. Oncol., 12: 671-678, 1994.

12. Lista, P., Brizzi, M. F., Avanzi, 0., Veglia, F., Resegotti, L., and

Pegoraro, L. Induction of proliferation of acute myeloblastic leukemia

(AML) cells with hemopoietic growth factors. Leuk. Res., 12: 441-447,

1988.

13. Gore, S. D., Weng, L. J., and Burke, P. J. Identification of growthfactor-responsive acute myelogenous leukemia based on factor-depen-dence for survival and proliferation. Leukemia, 8: 1854-1863, 1994.

14. Bloomfield, C. D., and de la Chapelle, A. Chromosome abnormal-ities in acute nonlymphocytic leukemia: clinical and biological signifi-cance. Semin. Oncol., 14: 372-383, 1987.

15. Geller, R. B., Burke, P. J., Karp, J. E., Humphrey, R. L., Braine,H. B., Tucker, R. W., Fox, M. 0., Zahurak, M., Morrell, L., Hall, K. L.,and Piantadosi, S. A two step timed sequential treatment for acutemyelocytic leukemia. Blood, 74: 1499-1506, 1989.

16. Gerdes, J. Ki-67 and other proliferation markers useful for immu-nohistological diagnostic and prognostic evaluations in human malig-nancies. Semin. Cancer Biol., 1: 199-206, 1990.

17. Gerdes, J., Lemke, H., Buisch, H., Wacker, H., Schwab, U., and

Stein, H. Cell cycle analysis of a cell proliferation-associated humannuclear antigen defined by the monoclonal antibody Ki-67. J. Immunol.,133: 1710-1715, 1984.

18. Kastan, M. B., Stone, K. D., and Civin, C. I. Nuclear oncoproteinexpression as a function of lineage, differentiation stage, and prolifer-ative status of normal human hematopoietic cells. Blood, 74: 1517-

1524, 1989.

19. Gore, S. D., Weng, L-J., and Burke, P. J. Validation of flow-cytometric determination of Ki67 expression as a measure of growthfactor response in acute myelogenous leukemia. Exp. Hematol., 21:

1702-1708, 1993.

20. Dolbeare, F., Kuo, W-L., Beisker, W., Vanderlaan, M., and Gray,J. W. Using monoclonal antibodies in bromodeoxyuridine-DNA analy-sis. Methods Cell Biol., 33: 227-234, 1990.

21. Miller, C. B., Zehnbauer, B. A., Piantadosi, S., Rowley, S. D., and

Jones, R. J. Correlation of occult clonogenic leukemia drug sensitivitywith relapse after autologous bone marrow transplantation. Blood, 78:

1125-1131, 1991.

22. Jones, R. J., Zuehlsdorf, M., Rowley, S. D., Hilton, J., Santos,G. W., Sensenbrenner, L. L., and Colvin, 0. M. Variability in 4-hy-droperoxycyclophosphamide activity during clinical purging for autol-ogous bone marrow transplantation. Blood, 70: 1490-1494, 1987.

23. Jones, R. J., Miller, C. B., Zehnbauer, B. A., Rowley, S. D., Colvin,

0. M., and Sensenbrenner, L. L. In vitro evaluation of combination drugpurging for autologous bone marrow transplantation. Bone MarrowTransplant., 5: 301-307, 1990.

24. Loken, M. R., Shah, V. 0., Dattilio, K. L., and Civin, C. I. Flowcytometric analysis of human bone marrow: II. Normal B lymphocytedevelopment. Blood, 70: 1316-1324, 1987.

25. Cheson, B. D., Cassileth, P. A., Head, D. R., Schiffer, C. A.,Bennett, J. M., Bloomfield, C. D., Brunning, R., Gale, R. P., Grever,M. R., Keating, M. J., Sawitsky, A., Stass, S., Weinstein, H., andWoods, W. G. Report of the National Cancer Institute-sponsored work-

shop on definitions of diagnosis and response in acute myeloid leuke-mia. J. Clin. Oncol., 8: 813-819, 1990.

26. Jones, R. J., Bedi, A., Miller, C. B., Kaufmann, S. H., Zilcha, M. S.,

Barber, J. P., Burke, P. J., and Gore, S. D. Inhibition of apoptosis: anovel mechanism of resistance to multiple cytotoxic agents in acutemyeloid leukemia. Exp. Hematol., 21: 1078, 1993.

27. Karp, J. E., and Burke, P. J. Enhancement of drug cytotoxicity ofrecruitment of leukemic myeloblasts with humoral stimulation. CancerRes., 36: 3600-3603, 1976.

28. Burke, P. J., Vaughan, W. P., Karp, J. E., and Sanford, P. C. Thecorrelation of maximal drug dose, tumor recruitment and sequencetiming with therapeutic advantage: schedule dependent toxicity of cy-tosine arabinoside. Med. Pediatr. Oncol., 10 (Suppl. 1): 201-208, 1982.

29. Bhalla, K., Tang, C., Ibrado, A. M., Grant, S., Tourkina, E., Hol-laday, C., Hughes, M., Mahoney, M. E., and Huang, Y. Granulocyte-macrophage colony-stimulating factor/interleukin-3 fusion protein(pIXY 321) enhances high-dose ara-C-induced programmed cell deathor apoptosis in human myeloid leukemia cells. Blood, 80: 2883-2890,1992.

30. Williams, G. T., Smith, C. A., Spooncer, E., Dexter, T. M., andTaylor, D. R. Haemopoietic colony stimulating factors promote cellsurvival by suppressing apoptosis. Nature (Lond.), 343: 76-79, 1990.

31. Miyashita, T., and Reed, J. C. bcl-2 Oncoprotein blocks chemother-apy-induced apoptosis in a human leukemia cell line. Blood, 82: 151-157, 1993.

32. Bedi, A., Vala, M. S., Barber, J. P., Sharkis, S. J., and Jones, R. J.BCR-ABL induces resistance to cytotoxic agents in chronic myeloid

leukemia (CML) through inhibition of apoptosis. Blood, 82 (Suppl. 1):327a, 1993.

33. Lowe, S. W., Ruley, H. E., Jacks, T., and Housman, D. E. p53-Dependent apoptosis modulates the cytotoxicity of anticancer agents.Cell, 74: 957-967, 1993.

34. McGahon, A., Bissonnette, R., Schmitt, M., Cotter, K. M., Green,D. R., and Cotter, T. G. BCR-ABL maintains resistance of chronicmyelogenous leukemia cells to apoptotic cell death. Blood, 83: 1179-1187, 1994.

35. Miyauchi, J., Kelleher, C. A., Wang, C., Minkin, S., and McCul-loch, E. A. Growth factors influence the sensitivity of leukemic stemcells to cytosine arabinoside in culture. Blood, 73: 1272-1278, 1989.

36. Park, L. S., Waldron, P. E., Friend, D., Sassenfeld, H. M., Price,V., Anderson, D., Cosman, D., Andrews, R. G., Bernstein, I. D., andUrdal, D. L. Interleukin-3, GM-CSF, and G-CSF receptor expressionon cell lines and primary leukemia cells: receptor heterogeneity andrelationship to growth factor responsiveness. Blood, 74: 56-65,

1989.

37. Budel, L. M., Touw, I. P., Delwel, R., Clark, S. C., and Lowenberg,B. Interleukin-3 and granulocyte-monocyte colony-stimulating factorreceptors on human acute myelocytic leukemia cells and relationship tothe proliferative response. Blood, 74: 565-571, 1989.

38. Kaufmann, S. H., Karp, J. E., Jones, R. J., Miller, C. B., Schneider,E., Zwelling, L. A., Cowan, K., Wendel, K., and Burke, P. J. Topoi-somerase II levels and drug sensitivity in adult acute myelogenousleukemia. Blood, 83: 517-530, 1994.

39. Lotem, J., and Sachs, L. Hematopoictic cytokines inhibit apoptosis

induced by transforming growth factor beta-i and cancer chemotherapycompounds in myeloid leukemia cells. Blood, 80: 1750-1757, i992.

40. Koistinen, P., Wang, C., Curtis, J. E., and McCulloch, E. A. Gran-ulocyte-macrophage colony-stimulating factor and interleukin-3 protect

leukemic blast cells from ara-C toxicity. Leukemia, 5: 789-795, 1991.

41. Gandhi, V., and Plunkett, W. Cell cycle-specific metabolism ofarabinosyl nucleosides in K562 human leukemia cells. Cancer Che-mother. Pharmacol., 31: 11-17, 1992.



1995;1:295-303. Clin Cancer Res S D Gore, L J Weng, R J Jones, et al.

leukemia.differentiation, and chemosensitivity of acute myeloid Impact of in vivo administration of interleukin 3 on proliferation,

Updated version

http://clincancerres.aacrjournals.org/content/1/3/295

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/1/3/295To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



impact of in vivo administration of interleukin 3 on ...€¦ · impact of in vivo administration...

Documents