Leukemia Research Vol. 18, No. 1, pp. 29-35, 1994. 0145-2126/94 $6.00 + .00 Printed in Great Britain. © 1993 Pergamon Press Ltd

FLOW CYTOMETRIC ANALYSIS OF CELL-SURFACE ANTIGEN EXPRESSIONS ON ACUTE MYELOID LEUKEMIA CELL

POPULATIONS ACCORDING TO THEIR CELL-SIZE

HIROSHI KAWADA, YUKINOBU ICHIKAWA, SHIGEKI WATANABE, TADAMI NAGAO and SHIGERU ARIMORI

The Fourth Department of Internal Medicine, Tokai University School of Medicine, Bohseidai, Isehara, Kanagawa, 259-11, Japan

(Received 22 June 1993. Revision accepted 24 August 1993)

Abstract--Acute myeloid leukemia (AML) cells which expanded from a single leukemic cell show certain degrees of morphological and biological heterogeneity. In the present study, we determined cell-surface antigen expressions (CD13, 33, 34 and 38, and HLA-DR) on AML cells based on their cell-size (large vs small cells) by flow cytometry. We found that the cell-surface antigens were more strongly expressed on the large leukemic cells than the small cells, regardless of FAB subtypes. Furthermore, our preliminary study demonstrated that AML patients who showed a relatively small difference in antigen expression between large and small leukemic cells had longer remission durations and survival periods, compared with those with a more prominent difference in antigen expression. Thus, the heterogeneity of AML cells determined by the combination of cell-surface antigen expressions and cell-size may be associated with clinically important biological behaviors.

Key words: Cell-surface antigens, cell-size, flow cytometry, acute myeloid leukemia, prog- nosis.

Introduction

ACUTE MYELOID LEUKEMIA (AML) cells which expanded from a single leukemic cell show various degrees of morphological and biological hetero- geneity. Certain cell-surface antigen expressions associate with the differentiation of leukemic cells [1, 2] and reflect the phenotypic heterogeneity of leukemic cells [3]. However, the biological and clini- cal significance of leukemic cell-size has not been of interest. The cell-size possibly associates with cell- surface marker expressions and may be an additional factor which relates to the heterogeneity of leukemic cells.

To elucidate possible differences of cell-surface marker expression according to leukemic cell-size,

Abbreviations: AML, acute myeloid leukemia; FAB, French-American-British; CR, complete remission; mAb, monoclonal antibody; FITC, fluorescein-isothiocyanate; PE, phycoerythrin; PBS, phosphate buffered saline; EDTA, ethylenediamine-tetraacetic acid; BSA, bovine serum albumin; MFI, mean fluorescence intensity; LS ratio, large/small cell ratio.

Correspondence to: Hiroshi Kawada, The Fourth Department of Internal Medicine, Tokai University School of Medicine, Bohseidai, Isehara, Kanagawa, 259-11, Japan.

therefore, we divided the leukemic cells obtained from AML patients at the time of diagnosis into large and small cell groups by using flow cytometry. Our results show that the cell-surface antigens are more prominently expressed on the large leukemic cells than small cells. Preliminary findings also suggest that these differences may correlate with biological features of AML cells.

29

Materials and Methods

Patients Twenty-three adult patients (10 males and 13 females,

mean age 48-years-old, range 20-79), who had been referred to Tokai University Hospital and were diagnosed as AML between December 1988 and August 1991, were included in the present study. They were classified accord- ing to the French-American-British (FAB) criteria [4, 5] and consisted of M1 (nine cases), M2 (seven cases), M3 (five cases), M4 (one case) and M5 (one case). All patients received BHAC-DMP regimen [6] as a remission induction chemotherapy. The regimen includes daily administrations of N4-behenoyl-l-fl-o-arabinofuranosylcytosine (BHAC, 170 mg/m2/day, d.i.v.), 6-mercaptopurine (6-MP, 70 mg/ m2/day, p.o.) and prednisolone (PSL, 20mg/m2/day, p.o.), and intermittent daunorubicin (DNR, 25 rag/m2/ day, bolus i.v.) for 10-14 days. DNR was routinely given on days 1 and 2 and was added on days 5, 6 and/or

30 H. KAWADAetal.

E "1

z

( . )

o O3

(D

0 3

2 0 0

A.

M

. • . . . . . |

B.

D.

M

/i L L J ~

' ' . ~ . ." . " " ' " I " . : . " . • • . ,

i ! :,!:::i!::~.: i : : ' i~: , .'~: , ,~J:i ::?

. . . . . : : . . . '..-~.~.:.'q:,~..: .. . . ' . . . . : : . ~ - ~ : . ~ , ' . : ..

P B l

• . . . • . i:.;.,~,i:;:;.,. ,,. . • ' . . . - : ; : I - . ' ~ , ' : . . . . . . ' . : ' ! ~ i ' ~ V , . .~ •

.i .i

..: • .,..: ~(~, ~.e::~ : . . . . . • ," ,; , ! : . i i . . .

- : . . . . , . . . ,~ i ' ,~ . : , . . . .

1 0 0 2 0 0 0 1 0 0 2 0 0

F o r w a r d S c a t t e r F o r w a r d S c a t t e r

FIG. 1. Gate settings based on cell-size. Cells were divided into (A) small and (B) large cell groups according to their mean channel number (M) of forward scatter which indicates the cell-size of leukemic cell. Each group of cells was gated (C and D) and then analyzed for the expression

of cell-surface markers.

9, depending on the individual patient's response to the therapy.

A complete remission (CR) was defined by normal test results from both bone marrow (less than 5% blasts) and peripheral blood. The overall CR rate was 82.6% (19 of 23 patients). Their CR durations ranged from 68 to 811+ days and survival periods were 1-842+ days.

Monoclonal antibodies (mAbs) Fluorescein-isothiocyanate(FITC)- or phycoerythrin-

(PE)-conjugated mAbs (Leu4 (CD3), My7 or LeuM7 (CD13), Leu12 (CD19). My9 or LeuM9 (CD33), Leu 17 (CD38) and anti-HLA-DR mAbs) were used in this study. An indirect immunofluorescent technique using FITC-lab- eled goat anti-mouse immunoglobulin F(ab')z (Becton- Dickinson Immunocytometry Systems, Mountain View, CA) was used for HPCA-1 (CD34) antigen determination. Leu-series mAbs, anti-HPCA-1 mAb, and anti-HLA-DR mAb were purchased from Becton-Dickinson Immuno- cytometry Systems, and My-series were from Coulter Immunology, Hialeah, FL.

Flow cytometric analysis Sample preparation. Heparinized peripheral blood (8

cases) or aspirated bone marrow samples (15 cases) were obtained before the treatment. The mean percentage of leukemic blasts determined morphologically was 85.2% in the samples (range 62-99%). Erythrocytes in the samples were lysed with hemolytic solution containing ammonium chloride (826 mg/dl), potassium bicarbonate (100 mg/dl) and tetrasodium ethylenediamine-tetraacetic acid (3.7 mg/

dl). After washing with phosphate buffered saline (PBS) supplemented with 0.05M disodium ethylenediamine- tetraacetic acid (EDTA) and 0.2% bovine serum albumin (BSA) cells were suspended in EDTA-BSA-PBS (5 x 105 cells/ml). Fifty microliters of the cell suspension were mixed with 1 ~g of FITC- and PE-mAbs, and the mixture was reacted at 4°C for 30 min in the dark. After washing with EDTA-BSA-PBS, the ceils were resuspended in 0.5 ml of EDTA-BSA-PBS, and were passed through a nylon mesh filter. More than 6000 cells were counted by a fluorescence activated cell sorter (FACSTAR ®, Becton- Dickinson). Appropriate gate settings were used to analyze the leukemic cell population.

Calculation of blast cells in the gated area. The per- centages of blast contained in the gated area were cal- culated by the following equation (equation (1)):

Blasts in the gated area (%) = Morphologically determined leukemic cells in the

sample (%) Total number of cells counted

X

Number of cells within the gated area"

To obtain more appropriate percentages of blast cells, we further excluded almost all lymphocytes (CD3+T-cells and CD19÷B-cells) contaminating within the gated area by using an additional formula.

Corrected percentage of blasts in the gated area = Blasts (%) in the gated area calculated by equation (1)

100 X

100 - (T-cells (%) + B-cells (%))"

FCM analysis of cell-surface antigen expressions

TABLE 1. CORRECTED PERCENTAGE OF POSITIVE CELLS WITHIN WHOLE LEUKEMIA CELLS

31

FAB Source % large subtypes Case of sample cells CD13 CD33 CD34 CD38 H L A - D R

M1 1 BM 38.7 97.4 100.0 0.3 88.4 0.4 2 PB 64.9 1.2 1.2 N.D. N.I. N.I. 3 BM 45.3 89.8 83.4 84.1 99.9 95.8 4 PB 45.1 2.5 27.6 N.D. 98.9 95.3 5 BM 46.5 29.8 27.7 63.8 100.0 99.4 6 PB 48.8 15.1 25.0 4.3 N.I. N.I. 7 BM 48.8 88.9 0.0 88.2 97.0 84.1 8 BM 48.8 29.5 87.2 3.1 N.I. N.I. 9 BM 51.9 58.6 87.5 N.D. 88.6 25.1

M ± S.D. 48.8 - 7.1 45.9 ± 38.6 48.8 ± 40.2 40.6 ± 42.5 95.5 ± 5.6 67.5 ± 43.3 M2 10 BM 33.4 70.4 N.D. N.D. 100.0 30.5

11 PB 46.9 37.8 25.0 N.D. 95.5 93.1 12 BM 50.3 13.8 37.6 96.6 97.4 92.5 13 BM 50.9 28.0 46.4 96.7 100.0 N.D. 14 PB 53.4 68.5 0.5 88.3 97.0 97.4 15 BM 49.0 64.6 89.8 41.6 100.0 98.8 16 PB 38.0 95.2 N .D. N .D. 98.5 96.1

M ± S.D. 46.0 ± 7.4 54.1 ± 28.4 49.6 ± 28.2 80.8 ± 26.5 98.3 ± 1.8 84.7 ± 26.7 M3 17 PB 54.9 36.1 31.7 N.D. 35.6 8.1

18 BM 47.0 45.6 33.6 1.4 12.3 1.3 19 BM 52.7 100.0 100.0 5.5 N.I. N.I. 20 BM 56.7 75.8 11.5 5.2 37.9 1.4 21 BM 56.0 92.9 77.5 3.8 N.I. N.I.

M ± S.D. 53.5 + 3.9 70.1 ± 28.3 50.9 ± 36.5 4.0 ± 1.9 a 28.6 ± 14.2 b 3.6 +- 3.9 c M4 22 BM 45.2 95.5 96.3 3.1 95.2 3.0 M5 23 PB 59.4 73.8 95.8 8.8 N.I. N.I.

Abbreviations: PB, peripheral blood; BM, bone marrow; N.D., not done; N.I. not included; M ± SD = mean ± one standard deviation. The p values of M3 vs M2 (a), M3 vs M1 or M2 (b) and M3 vs M1 (c) are less than 0.05.

The corrected % of blasts in the gated area was higher than 90% in all the patients studied. In addition, patients with mixed lineage leukemia including CD19*AML [7, 8] were not included in the present study.

Cell-surface marker expression. Percentages of cell-sur- face marker-positive blast cells were calculated by a similar equation, which also excluded lymphocyte contaminations.

Corrected percentage of positive cells -- positive cells in the gated area (%)

100 x

100 - (T-cells (%) + B-cells (%)) '

When the corrected percentage of positive cells was higher than 20%, the leukemic blasts were considered as positive for each antigen.

The sum of T- and B-cells within the gated area was more than 10% in six cases, and their results were excluded from the evaluations for CD38 and H L A - D R antigen expressions, since certain numbers of lymphocytes were shown to he positive for the antigens.

Leukemic cell analysis according to the cell-size. The cells within the gated area were divided into large and small cell groups by their mean channel number (mean cell-size) of forward scatter (Fig. 1). The mean percentage of large cells within the gated area was 49.2 ± 7.0% (range 33.4-64.9%) (Table 1). The corrected percentage of positive cells and the mean fluorescence intensity (MFI) of cell-suface markers, which were considered as positive, were com- pared between large and small cell groups in each patient.

As an indicator showing which cell group more strongly expresses cell-surface antigens, large and small cell ratios (LS ratios) for the corrected percentage of positive cells or MFI corresponding to each cell-surface marker were further calculated in each patient.

Corrected percentage of positive cells or MFI of cell-surface antigen in large cell group

Corrected percentage of positive cells or MFI of cell-surface antigen in small cell group

According to the median of each LS ratio, patients were divided into two groups with high and low LS ratios.

For statistical analyses, the Wilcoxon matched-pairs signed-ranks test, Wilcoxon rank sum test or Kaplan-Meier method were used in the present study.

Resu l t s

Express ions of ce l l -surface m a r k e r on the b las t cells ( co r rec t ed % of pos i t ive cells) wi th in the who le ga t ed a rea were d i f fe ren t a m o n g the F A B sub types (Tab le 1). CD34 an t igen express ions were h igher on M2 blasts than M3 blasts ( p < 0.05). The p e r c e n t a g e s of M1 and M2 blasts which exp re s sed C D 3 8 an t igen were h igher than M3 blas ts ( p < 0.05). H L A - D R an t igen was exp re s sed at h igher ra tes on M1 blas ts

32 H. KAWADA et al.

I00-

t~ 50

-.~

O

P < 0.01 i i

, s

CD13

P < 0.01 r'-----i

I

L S

CD33

\-

P < O.Ol I I

T[

. . . . i

\

I r I r

L S L S

CD34 CD38

P< 0.01 ]

" " }

_L

t 1 L S

H L A - DR

e -

g

500

400

300 •

200 -

100.

FIG. 2. Comparison of the corrected percentage of positive cells for each cell-surface marker between large (L) and small (S) cells. Most of the markers were highly expressed

on the large AML cells compared with the small cells.

P < 005 I I

I r

L S CD13

100

50

P < 0.01 I

P< 0.05 500 P 0.01 1200 ~ i i

1100

I000 400

90o

8oo

600 ~ .

500 i 200-

400 i 7 ! i

300 ~ 100 "

200 ~

100 ~---~.

8oo iii t 500-

400-

300-

200-

100-

I I ' i " I I I

L S L S L S

CD33 CD34 CD38

FIG. 3. Comparison of the mean fluorescence intensity of each cell-surface marker between large (L) and small (S) cells. As observed in Fig. 2, each marker was strongly expressed on the large AML cells in most of the patients.

P< 0.01

\

I I

L S

HLA- DR

than M3 blasts (p < 0.05). However, the data of CD38 or H L A - D R antigen were based on only three patients with M3 subtype. There was no significant difference in the CD13 and CD33 expressions among the FAB subtypes. Additionally, no significant dif- ference was found in the remission rates, CR dur- ations and survival periods between patients' groups

whose blasts were positive and negative for each cell- surface marker.

When large and small blasts were compared, CD13 (74.6 - 25.3% vs 60.0 -+ 27.7%), CD33 (69.9 --+ 30.3% vs 54.4 + 33.1%), CD38 (91.5 +--18.9% vs 8 7 . 0 - 23.6%) and HLA-DR-posit ive cells (87.2--- 24.6% vs 79.1---30.0%) were more frequently

v G g:

2

(A) I O0

9 0 ~ 80 7 0

60.

go.

40-

30.

~0.

I0 .

0

(c) I00

90 ~ . ,

80

?O

60 SO

40 30

20

I0

0

FCM analysis of cell-surface antigen expressions

in cemlsslon

(2)

1 4~o

Days

P < O.OS

i ( I )

eO0 .6o

alive

1

P < 0.001

(2)

! (1)

I00

90-

80-

~ 7 0

~ 60- g: 50-

i~ ,o_

wE 30- t12

20-

200

I0-

(D) IO0

90.

8 0

~ 70-

6o. 1 o~ 5o-

"~ 40-

~ 30-

20-

I0-

0 400 600 800

Days

FIG. 4 (A) (B). Comparison of

( I )

2~o 3~o Days

33

! in femlssioo

: (2)

P < O.OOI

400 SO0

! alive

1

L (2)

P<001 (1)

2;0 4;0 ~;o .;o Days

remission durations between patients with high large/small cell (LS) ratio (1) and those with low LS ratio (2) for the mean fluorescence intensity of CD13 antigen ((A): (1) n = 8, (2) n = 6) and for the corrected percentage of positive cells for CD38 antigen ((B): (1) n = 5, (2) n = 5). (C) (D). Comparison of survival periods between patients with high LS ratio (1) and those with low LS ratio (2) for the mean fluorescence intensity of CD33 antigen ((C): (1) n = 9, (2) n = 8) and for the corrected percentage of positive cells for HLA-DR antigen ((D): (1) n = 6, (2) n -- 5). Three dots represent

one patient who is in remission or alive.

detected in the large leukemic blasts than the small cell population (p < 0.01) (Fig. 2). However, CD34- positive cells did not differ between the two leukemic blasts (83.5 + 15.8% vs 76.4 + 24.6%).

We further compared the MFI of each antigen between the two groups of cells. The MFIs of CD13 (134.5 +- 92.3 vs 121.3-+ 78.1, p < 0 . 0 5 ) , CD33 (58.2-+ 28.1 vs 50.4 + 27.0, p <0 .01) , CD34 (278.3 + 257.9 vs 233.1 + 190.3, p < 0.05), CD38 (225.5 + 129.5 vs 164.5 -+ 91.4, p < 0.01) or HLA- DR (430.0 +2 8 0 .0 vs 285.6-+ 165.4, p < 0 . 0 1 ) were also found to be high for the large cell popu- lation compared with the small cell population (Fig. 3).

In the following analysis, LS ratios for the cor- rected percentage of positive cells or their MFIs were

compared with patients' sexes, ages, FAB subtypes (M1, M2, M3), responses to the induction chemo- therapy and prognosis. There was no correlation of the LS ratios with both patients' sexes and ages. The LS ratio for MFI of CD13 antigen, however, was higher in M1 blasts than M3 blasts (1.12 -+ 0.14 vs 0.92 +-- 0.12, p < 0.05). Patients with a low LS ratio for MFI of CD13 antigen (p < 0.05) or corrected percentage of CD38-positive cells ( p < 0 . 0 0 1 ) exhibited longer remission durations than those with a high LS ratio for each parameter (Fig. 4(A), (B)). Similarly, patients with low LS ratios for MFI of CD33 antigen (p < 0.001) or corrected percentage of HLA-DR-positive cells (p < 0.01) showed longer survival periods than those with high LS ratios for the respective parameters (Fig. 4(C), (D)).

34 H. KAWADA et al.

Discussion

CD 13 and CD33 antigens are frequently expressed on AML cells (FAB subtypes M1-M5), while the expressions of CD34 and HLA-DR antigens are infrequent in M3 blasts [9]. CD38 antigens are gen- erally expressed on myeloid cells after CD34 antigen expression, and both antigens decrease during further myeloid cell differentiation [10]. Our results agree with these previous reports and support the concept that A M L M3 is on more mature stages of myeloid differentiation as compared with A M L M1 and M2 [4].

In the present study, we first assessed cell-surface marker expressions of A M L cells according to their cell-size. Our results clearly showed that the cell-sur- face antigens examined in the present study were more strongly expressed on large rather than small AML cells in most cases, regardless of their FAB subtypes. Associations of tumor cell-size with other charac- teristics of tumor cells including cell-surface marker expressions have been rarely studied, but one report described different expressions of HLA-DR antigen between large and small tumor cells from malignant lymphoma patients (follicular mixed-cell type) [11].

Since several studies demonstrated relationships of cell-cycle with cell-size and/or cell-surface marker expression [12-14], our results might reflect the dif- ferences in cell-cycle of leukemic cell populations. However, A M L cells in both S-phase and G 2 + M phase are very few in v i vo [15]. Therefore, the dif- ferences in cell-surface marker expressions between small and large A M L cells were considered to show additional characteristics of A M L cells, although the differences were relatively small and showed a broad range of variation among patients.

Several cell-surface marker expressions of leu- kemia cells are considered as prognostic factors of leukemic patients [7, 16, 17]. Although the number of patients studied in this report was small, we could not find any associations of cell-surface marker expressions on whole blast cells with patients' prog- nosis. We found that the patients whose large leu- kemic cells showed higher CD13-, CD33-, CD38 or HLA-DR antigen expression, compared with their small cell population, had short remission durations or survival periods. The differences in prognosis were still significant after exclusion of AML M3 cases from the present results (data not shown).

Thus, our preliminary results suggest that the dif- ference in cell-surface antigen expressions of AML cells based on their cell-size may be associated with clinically important biological behaviors of leukemic cells. However, studies including more patients will be required to confirm the present results.

Acknowledgements--The authors thank Miyoko Yoshida, Nobumasa Kobayashi, Mieko Takei, Yukari Arahira and Dr Shuji Yonekura for their assistance.

References

1. Terstappen L. W. M. M., Safford M., Unterhalt M., K6nemann S., Zurlutter K., Piechotka K., Drescher M., Aui C., Biichner T., Hiddemann W. & Wfrmann B. (1992) Flow cytometric characterization of acute myeloid leukemia: IV. Comparison to the differenti- ation pathway of normal hematopoietic progenitor cells. Leukemia 6, 993.

2. Griffin J. D., Mayer R. J., Weinstein H. J., Rosenthal D. S., Coral F. S., Beveridge R. P. & Schlossmann S. F. (1983) Surface marker analysis of acute myeloblastic leukemia: identification of differentiation-associated phenotypes. Blood 62, 557.

3. Terstappen L. W. M. M., Safford M., KOnemann S., Loken M. R., Zurlutter K., Biichner T., Hiddemann W. & W6rmann B. (1991) Flow cytometric charac- terization of acute myeloid leukemia. Part II. Pheno- typic heterogeneity at diagnosis. Leukemia 5, 757.

4. Bennett J. M., Catovsky D., Daniel M.-T., Flandrin G., Galton D. A. G., Gralnick H. R. & Sultan C. (1976) Proposals for the classification of the acute leukemia. A report of the French-American-British Cooperative Group. Br. J. Haemat. 33, 451.

5. Bennett J. M., Catovsky D., Daniel M.-T., Flandrin G., Galton D. A. G., Gralnick H. R. & Sultan C. (1985) Proposed revised criteria for the classification of acute myeloid leukemia. Ann. Intern. Med. 103, 626.

6. Kato S., Kawashima Y., Ohono R. & Yamada Y. (1983) Treatment of adult acute nonlymphocytic leu- kemia with BHAC-DMP therapy. Acta Haemat. Jpn

"4, 1077. 7. Ball E. D., Davis R. B., Griffin J. D., Mayer R. J.,

Davey F. R., Arthur D. C., Wurster-Hill D., Noll W., Elghetany M. T., Allen S. L., Rai K., Lee E. J., Schiller C. A. & Bloomfield C. D. (1991) Prognostic value of lymphocytic surface markers in acute myeloid leukemia. Blood 77, 2242.

8. Kita K., Nakase K., Miwa H., Masuya M., Nishii K., Morita N., Takakura N., Otsuji A., Shirakawa S., Ueda T., Nasu K., Kyo T., Dohy H. & Kamada N. (1992) Phenotypical characteristics of acute myelocytic leukemia associated with the t(8;21)(q22;q22) chromo- somal abnormality: frequent expression of immature B-cell antigen CD19 together with stem cell antigen CD34. Blood 80, 470.

9. Geller R. B., Zahurak M., Hurwitz C. A., Burke P. J., Karp J. E., Piantadosi S. & Civin C. I. (1990) Prognostic importance of immunophenotyping in adults with acute myelocytic leukaemia: the sig- nificance of the stem-cell glycoprotein CD34 (Myl0). Br. J. Haemat. 76, 340. Terstappen L. W. M. M., Huang S., Safford M., Lansdorp P. M. & Loken M. R. (1991) Sequential generations of hematopoietic colonies derived from single nonlineage-committed CD34÷CD38 - progenitor cells. Blood 77, 1218. Ratech H. (1990) HLA-DR expression in B-cell non- Hodgkin's malignant lymphomas: a multiparameter flow cytometry study. Hum. Path. 21, 1275.

10.

11.

FCM analysis of cell-surface antigen expressions 35

12. Gavosto F., Pileri A., Bachi C. & Pegoraro L. (1964) Proliferation and maturation defect in acute leukaemia cells. Nature 203, 92.

13. Campos L., Ffrench M. & Guyotat D. (1990) Myeloid differentiation antigens identify leukemic cell sub- populations with different cell cycle characteristics. Leukemia 4, 60.

14. Takase K., Okawa H., Minato K. & Yata J. (1988) Cell cycle-related expression of surface antigens on myelomonocytic leukemia cells. Leukemia Res. 12, 583.

15. Powell B. L., Gregory B. W., Evans J. K., White J. C., Lyerly E. S., Chorley H. M., Russell G. B. & Capizzi R. L. (1991) Leukapheresis induced changes

16.

17.

in cell cycle distribution and nucleoside transports in patients with untreated acute myeloid leukemia. Leu- kemia 5, 1037. Schwarzinger I., Valet P., KOler U., Marosi C., Schneider B., Haas O., Knapp W., Lechner K. & Bettelheim P. (1990) Prognostic significance of surface marker expression on blasts of patients with de novo acute myeloblastic leukemia. J. clin. Oncol. 8, 423. Campos L., Guyotat D., Archimbaud E., Devaux Y., Trelle D., Larese A., Maupas J., Gentilhomme O., Ehrsam A. & Fiere D. (1989) Surface marker expression in adult acute myeloid leukaemia: cor- relations with initial characteristics, morphology and response to the therapy. Br. J. Haemat. 72, 161.