content by increasing the relative Gall fraction in sam- ples from aspirates as ..... the DNA content distribution during the course of dis- ease in the single patients.
Cytometry 4:238-243 (1983)
0 1983 Alan R. Liss, Inc.
Flow Cytometric DNA Content in Myelodysplastic Syndromes’ Carlomaurizio Montecucco, Albedo Riccardi, Egidio Traversi, Marco Danova, Giovanni Ucci, Giuliano Mazzini, Paolo Giordano Istituto di Patologia Medica I (C.M., A.R., E.T., M.D., G.U.), Centro di Studio Per I’Istochimica del CNR and Istituto di Anatomia Comparata (G.M., P.G.),Universita di Pavia, Pavia, Italy Received for publication February 23, 1983; accepted May 31, 1983.
DNA flow cytometric analysis of unfixed bone marrow cells stained with propidium iodide was carried out in 33 patients with untreated primary myelodysplastic syndromes. Patients with stable clinical course for up to 3 years had higher fractions of cells in S and G 2 phases (22.7 & 12.4% and 12 3.6%) than those who developed acute leukemia andlor died early in the course of disease (14.4 & 8.5% and 6.6 f 4%). Median survival was more than 36 mo in patients with S+G2 cell fraction higher than 24%, and 14 mo in the remaining 16 patients with lower values (P < 0.01). Analyses repeated after 3-24 mo showed no major changes in cell proliferation
Myelodysplastic syndromes are primary acquired hematologic disorders characterized by refractory cytopenia of one or more hemopoietic cell lines, due to defective hemopoiesis (3). These conditions may progress to overt leukemia, may remain stable for many years, or may lead to death due to bone marrow failure (9,11,14,22).Based on cell morphology of both bone marrow and peripheral blood, a clinical classification has been recently proposed that seems endowed with prognostic value (3). However, a n universal agreement is very far from being reached on this topic (24). Studies, carried out by cytogenetic and autoradiographic techniques, showed that both the presence of chromosome abnormalities (19)and the reduction of proliferative activity in bone marrow cells (8,231 are correlated with a bad prognosis and a high risk of developing acute leukemia. Since some information on both cell proliferation pattern and cell ploidy (2) can be achieved by flow cytometry, we studied, in this way, bone marrow cells on 33 patients with primary myelodysplastic syndromes. The results were correlated with the clinical diagnosis according to the French-American-British (FAB) classification (3), the transformation into acute leukemia, and the patient survival.
pattern in ten out of 11 patients. The remaining patient had sharp decrease in S and G2 cell fraction 3 mo before the transition into acute leukemia. The DNA index (DI) of bone marrow cells was calculated to assess ploidy. However, comparative evaluation of cytologic, cytogenetic, and flow cytometric data suggest that, under our experimental conditions, the DI may be influenced by factors such as the degree of chromatin compactness. Key terms: Myelodysplastic syndromes, refractory anemia, preleukemia, DNA flow cytometry
PATIENTS AND METHODS Patients Thirty-three patients who were diagnosed as having myelodysplastic syndrome were studied. Patients with previous exposure to myelotoxic drugs or chemicals and radiations were excluded. Median age was 65 (range 1483). Examinations of both marrow aspirate and biopsy section, as well as routine cytogenetic analysis, were performed in all cases. Based on the guidelines proposed by the FAB Study Group (3), patients were categorized in the following groups. Refractory anemia (RA) was diagnosed in six patients. They had cellular marrow with increased erythro: granulocytic ratio and evidence of dyserythropoiesis.
’Research supported by CNR (Consiglio Nazionale delle Ricerche), special project “Control of Neoplastic Growth,” grant nos. 82.00219.96 and 204.212.96.93251. Presented in part at the Combined International Conference on Analytical Cytology and Cytometry, IX and the VIth International Symposium on Flow Cytometry, Schloss Elmau, West Germany, 18-23 October, 1982. Address reprint requests to Dr. Carlomaurizio Montecucco, Istituto di Patologia Medica I, Policlinico S. Matteo, 27100 Pavia, Italy.
DNA FLOW CYTOMETRY IN MYELODYSPLASIAS
Myeloblasts were less than 5% of bone marrow nucleated cells in all cases. Ring sideroblasts were almost absent. Refractory anemia with ring sideroblasts (RARS) was diagnosed in six patients who had cytologic and histologic features similar to those seen in RA. The distinctive characteristic was the presence of ring sideroblasts (more than 15% of bone marrow cells) with Prussian Blue staining. Refractory anemia with excess of blasts (RAEB) was diagnosed in 11 patients who had 10-20% myeloblasts in bone marrow aspirates. Auer rods were never found. Ring sideroblasts did not exceed 5% in any patient. Circulating immature granulocytic cells were occasionally observed on smears made from the buffy coat. Chronic myelomonocytic leukemia (CMML) was diagnosed in four patients with bone marrow features similar to those of RAEB but showing increased absolute number of monocytes (more than lOOO/pl) in peripheral blood. RAEB in transformation (tRAEB) was diagnosed in six patients who had marrow myeloblast percentage ranging from 20% to 30%. Immature granulocytic cells were always less than 10% of circulating leukocytes. One patient had Auer rods in a few marrow myeloblasts. A bone marrow reexamination, carried out two months later, showed no significant change in myeloblast percentage in this patient. Erythroblastic hyperplasia with prominent dyserythropoiesis was found in both examinations. An increased percentage of megakaryocytes, most of them showing the features of micromegakaryocytes (3,22) was also observed. All these data lead us to diagnose myelodysplasia, rather than early myeloblastic leukemia, in this case too. All patients were followed for a period of a t least 36 mo or until death. Fifteen patients are still alive. Seven patients (two from sepsis, two from hemorrhage, two from transfusion siderosis and one from cerebral thrombosis) died within 18 months, and 11patients developed acute leukemia (3)between 3 and 27 mo (Table 1). Median survival after leukemic transformation was 4.3 mo (range: 1 day-12 months). All patients were studied at their first admission to our institute before any treatment (blood transfusions, desferrioxamine, androgens, prednisone and, in three cases, low dose chemotherapy) was started. Eleven patients were also studied serially during the course of the disease. All the five patients studied during the acute leukemic phase had more than 50% blast cells in the bone marrow.
DNA Flow Cytometry Flow cytometric analysis was performed in all cases on bone marrow samples obtained by needle aspiration. In marrow aspirates, variable degrees of admixture between peripheral blood and marrow cells are usually seen. This may affect flow cytometric analysis of DNA content by increasing the relative Gall fraction in samples from aspirates a s compared to biopsies (7,12). In order to eliminate contaminating blood cells, only mar-
239
row particles were collected for cytometric analysis in our patients. All patients in this study had enough particles for both cytologic and flow cytometric examination. Three to five particles (depending on their own size) were placed on a sloping slide and gently washed by saline. They were thereafter suspended in 2 ml of phosphate-buffered saline and then drawn through needles of decreasing diameter to obtain cell suspension. Whole peripheral blood of healthy subjects was used a s diploid reference standard. DNA staining was carried out according to the method of Krishan (15), slightly modified (21). A few drops of peripheral blood or bone marrow cell suspension were stained with 2-4 ml A grade propidium iodide (PI, Calbiochem-Behring Corp., San Diego, CA) 50 pg/ml in hypotonic (0.1%) solution of NA citrate. The nonionic detergent Triton X 100 (Hoechst AG, West Germany) 0.1% was added in order to enhance the lysis of cell membrane. The final concentration of cells in the staining solution was kept in the range of 1-2 x 105/ml. A 10-min staining time a t 4°C proved to provide the best histogram resolution. Bone marrow samples were filtered through 50 pm-pore nylon filter before measurement was performed by a 4800 A Cytofluorograph (Bio Physics Systems, Inc., Mahopac, NY). Normal blood was used to calibrate the instrument. Thereafter, patient sample was analyzed under the same instrumental conditions. A further measurement of normal blood was finally carried out to confirm that these conditions were kept constant. A total of 50,000 cells were measured for each peripheral blood and bone marrow sample. The values of CV (5)ranged from 2.6 to 5.9 (median 4.5) for the control samples, and from 4.3 to 9.4 (median 6.0) for the Go/l peak of the patient bone marrow samples. In 32 patients with normal distribution, cell cycle phase analysis was carried out according to Fried’s model (10) adapted to a Honeywell 6040 computer. According to Barlogie et al. (l), both the presence of extra peaks in DNA profile and of a skewness of the Go/l peak were considered distribution abnormalities suggesting aneuploid? (type A and C aneuploidy). However, a tetraploid cell fraction exceeding 15% (type B aneuploidy of Barlogie et al.) (1)was not ascribed to the presence of a n aneuploid cell clone in our patients. Polychromatic erythroblasts are, in fact, frequently seen to be arrested in G2 phase of the cell cycle, prior to mitosis, in refractory anemias (17,26).The shift along the abscissa of the whole population (Type D aneuploidy) was defined by the evaluation of the PI-DNA index (DI = modal channel of bone marrow celldmodal channel of diploid reference standard). Normal range of DI was defined by the mean 2 SD (0.99 f 0.056) of the values obtained from 25 bone marrow samples of control subjects who had no hematologic disorder. No significant difference was found between DI values obtained before and after addition of RNase (Sigma, St. Louis, MO; 50 d m l for ten min at room temperature) (27) in 16 control subjects and in 11 patients with myelodysplastic syndromes. Such a treatment had been proved to be enough to produce the
240
MONTECUCCO ET AL
maximal reduction in fluorescence intensity (25) by removing double-stranded RNA (27).
RA
Statistical Analysis Statistical analysis was performed using Student's t test for the assessment of the difference between means. Life tables for two groups of patients were drawn up and the difference in survival was assessed by the log-rank test (20).
> 50
P)
251
OJ , 0
Table 1 shows that the fraction of cells in S and G2 phases is high in patients without excess of blasts (RA and RARS) who usually have a good prognosis. On the contrary S and Gz cell percentage was distinctly lower in patients showing increased amounts of marrow myeloblasts. Regardless of clinical diagnosis the percentage of S and G2 cells was higher in patients with longlasting stable disease (mean SD 22.7 *12.4% and 12.7 f 3.6%) than those who died within 36 mo following the study (14.4 8.5% and 6.6 5 4%) (P < 0.01). Patients who developed acute leukemia differed from long survivors in the percentage of G2 cells, in S G2 cell fraction and S:G2 cell ratio (Table 2). Accordingly, patients with fraction of S + G2 cells higher than 24.2% (median value of the whole series) showed longer survival than those with lower values. The former population grouped mostly patients with RA and RARS, while the latter grouped patients with more malignant conditions such as CMML and tRAEB (Fig. 1).Patients with RAEB were the most heterogeneous group as far as the percentage of cell in S and G2 phases is concerned. Of the four patients with high percentage of S and G2 phase cells, three are still alive and one died from acute leukemia. Of the seven patients with low values, five died within 36 mo (two from leukemia). Repeated examinations of six patients who did not enter acute leukemic phase showed no significant changes in S + G2 cell fraction during a period of 3-24 mo. Furthermore, the fraction of cells in S and G2 phase did not change in four out of five patients studied both in preleukemic and terminal acute leukemic phase. The remaining patient showed a reduction in cell proliferation, becoming evident three mo before acute leukemia was morphologically detectable (Fig. 2).
1
=
2
CMML tRAEB
I
I
I
24
36
I
12
=0 =3
=
L
months
FIG.1. Survival curves of two groups of patients separated by flow cytometric analysis of DNA of bone marrow cells. The S + G2 cut off value was 24.2%, i.e., the median value of the whole series of patients. (abbreviations a s in Table 1).
*
*
50j
25
+
Ploidy Analysis
-
1
=
RA RARS
c.
5
CMML =
t RAEB
L
n
RESULTS Cell Cycle Phase Analysis
=L
RARS= 6
0
v)
6
12
18
mo.s after diagnosis
24
mo.s 27 before acute 6 leukemia 3 0 FIG.2. Changes in S + G2 cell fraction found during the course of disease in six patients who did not enter acute leukemia (upper part) and in five patients who entered them within a period of 4-27 mo.
creased DI had diploid or pseudodiploid (1 case) karyotype, and all six patients with DI lower than 0.935 showed only diploid mitoses. Six out of eight patients with high DI values had preleukemic disease, while none of the six patients with low DI developed acute leukemia. All patients with high DI values had excess of blasts and seven of these had a reduced erythroblast percentage (lower than 20%). On the contrary, only one out of six patients with DI lower than normal had some increase in marrow myeloblast percentage, and four patients had erythroblast percentage exceeding 50% a t time of study.
Except for one patient who showed a skewness of the Gall peak, all patients had unimodal distribution of Go/l cells and did not show any extrapeak on the DNA histogram. The PI-DNA index (DI)was within normal range in 17 patients, higher in ten, and lower in six. Figure 3 shows the relationship of DI values with cytogenetic data and clinical outcome in 25 patients with available DISCUSSION chromosome analysis. Evidence of aneuploid metaFlow cytometric analysis of DNA content of bone marphases was obtained in four out of eight patients with DI higher than 1.045 and in two out of 11 patients with row cells in myelodysplastic syndromes allowed us to DI falling in the normal range. Four patients with in- separate two groups of patients with different clinical
241
DNA FLOW CYTOMETRY I N MYELODYSPLASIAS
Table 1 Clinical Course and D N A Flow Cytometric Data Related to Clinical Diagnosis in 33 Patients With Various Myelodysplastic Syndromes Clinical diagnosis" RA
No. of patients
6
No. of deaths
No. of leukemias
1
0
Median DNA values (range) G2-phase (76) S-phase (%)
16.8 9.8-45.7) 27.3 (12.2-35.6) 12.4 ( 5.5-37.3) 8.6 ( 3.7-29.7) 15.4' (10.7-28.5) 14.7 ( 3.7-45.7) (
RARS
6
1
1
RAEB
11
6
3
CMML
4
4
2
tRAEB
6
6
5
33
18
11
Total
11.8 (9.7-17.4) 14.7 (7.5-20.0) 9.5 (2.1-15.2) 4.5 (0.1-7.8 ) 4.5b (2.4-9.3 ) 9.6 (0.1-20.0)
"RA = refractory anemia; RARS = RA with ring sidroblasts; RAEB = RA with excess of blasts; CMML = chronic myelomonocytic leukemia; tRAEB = RAEB in transformation. bThe data concern five patients only; the remaining patient had irregular pattern of DNA distribution, hampering cell cycle phase analysis.
Table 2 D N A Flow Cytometric Data Related to Clinical Course and Survival: Comparison With Data Obtained in 25 Normal Subjects
Clinical Course
> 3 year survival (No. of patients: 15) < 3 year survival (No. of patients: 17) Transition to leukemia (No. of patients: 11) Controls (No. of patients: 25)
S-phase (8)
17.3 (6.9-45.7) 12.3b (3.7-30.4) 12.6 (7.0-30.4) 15.6 (7.7-30.9)
Cell cycle phase analysis median values (range) Gz-phase (%) S + Gz (%)
S:G2 ratio
32.0 (16.9-63.1) 19.4" ( 5.5-41.7) 17.1b ( 7.7-41.7) 21.9 ( 8.2-38.4)
1.3 (0.6-3.8 ) 2.2 (0.4-27.0) 2.7' (1.3-15.0) 3.1 (0.9-15.4)
11.8 (7.5-20.0) 6.7" (0.5-15.2) 5.8" (0.5-11.3) 6.4 (0.5-11.6)
aP < 0.01. b P < 0.02. 'P < 0.05 (as compared with the values found in the first group).
characteristics and survival. Patients with high fraction of S and G2 cells had, as a rule, stable clinical course up to 3 yr. However, low percentages of these cells proved to be correlated with a n elevated propensity to develop acute leukemia and with short survival. The percentage of cells in S and G2 phases may reflect both the degree of proliferative activity of the overall bone marrow cell population and the pathological presence of cells arrested during their transit throughout the cell cycle. The latter condition seems to occur in erythroblasts as a sign of ineffective erythropoiesis (26). Thus i t may take place in increasing S and G2 cell fractions in disorders (such as RA and RARS) characterized by red cell hyperplasia with prominent features of dyserythropoiesis (3) and ineffective erythropoiesis (4).However, many patients showing an excess of blasts had low fractions of S and G2 cells, as is usually found in acute leukemia (2). Both
the low erythrogranulocytic ratio (3) and the reduction in proliferative activity of dysplastic granulocyte precursors (17,23) may account for this finding. In clinical practice, simple morphological considerations (3) allow entities with different prognosis to be separated. As is shown in Table 1, myelodysplastic syndromes without excess of blasts had mild clinical course, while patients with increased blast percentage in bone marrow (chiefly CMML and tRAEB) had very poor prognosis, being, in most cases, in a preleukemic condition. In this regard, flow cytometry seems to provide a further rationale for this classification. Our series is too small to ascertain whether DNA distribution analysis is endowed with prognostic meaning within single clinical subsets. Some evidence exists, however, that insights on prognosis may be obtained in patients with RAEB, who proved to be a quite heterogeneous group as far as both
242
MONTECUCCO ET AL
normal DI in our series showed unimodal Gail cell distribution without any extra peak or skewness. This seems to exclude the mixture of subpopulations with different DNA stem lines. Therefore, factors other than 1.1 the changes in chromosome pattern may influence the DI under our experimental conditions. These factors may X be related either to methodological aspects, such as speed -L a, of analysis and dye/cell ratio, or to changes in cellular U .-C dye uptake. Both the staining time (10 min) and flow 0 A < 1.0 2 rate of cell during measurement (500/sec) were kept as z n constant as possible. The duration of analysis was about 0 I 100 sec to reach a total cell count of 50,000 in all sam& ples. Furthermore, dye/cell ratio was set to allow all ow available binding sites of DNA to be saturated by PI. In 0.9 fact, the calculated value for R (R = pM dye/pM DNAP) (16) was always greater than 10. Thus, methodological factors do not seem responsible for changes in DI in our patients. Changes in dye uptake due to osmotic s R A 1 PL contro M D S resistance and double-stranded RNA content were circumvented by adding both Triton X 100 and RNase to FIG. 3. Propidium iodide-DNA index (DI) of bone marrow cells in 25 the staining solution (25). However, the degree of chronormal subjects and in 25 patients with myelodysplastic syndromes (MDS) with available cytogenetic data ( 0 ,patients with diploid or matin compactness may also affect dye uptake in cells pseudodiploid karyotype; A , patients with cytogenetic evidence of with identical DNA content (16,18,25). It is possible that aneuploidy; sRA, patients with stable disease up to 36 mo; PL, patients differences in the staining properties of chromatin in who developed acute leukemia). situ are responsible for the discrepancies between cytogenetic and cytometric data. In this regard, we stress the presence of large fractions of erythroblasts in samcell proliferation pattern and clinical course are ples with low DI and of immature granulocyte precurconcerned. sors in those with high DI. This is not surprising, since Repeated studies showed a remarkable constancy in erythroblasts have a compact chromatin structure, while the DNA content distribution during the course of dis- myeloblasts and undifferentiated blast cells have, as a ease in the single patients. In most patients, evolution rule, a n uncondensed chromatin pattern leading to an into overt leukemia without changes in cell proliferation easier intercalation by phenantridinic dyes. pattern also occurred. These findings are in accordance LITERATURE CITED with those of Barlogie et al. (2), who did not find any difference in DNA distribution between oligoblastic and 1. Barlogie B, Hittelman W, Spitzer G, Trujillo J , Hart J, Smallwood L, Drewinko B: Correlation of DNA distribution abnormalities overt acute leukemia. This is quite surprising, as it with cytogenetic findings in human adult acute leukemia and suggests that both leukemic blasts and partially differlymphoma. Cancer Res 37:4400-4407, 1977. entiating cells of preleukemic phase have similar prolif2. Barlogie B, Latreille J, Freireich E, Mellard D, Meistrich M, erative characteristics. These data are reminiscent of Andreef M: Characterization of hematologic malignancies by flow those seen in patients with acute leukemia at remission. cytometry. Blood Cells 6:719-744, 1980. 3. Bennet JM,Catovsky D, Daniel MT, Flandrin G, Galton DAG, In some of these patients, a reduction of S-phase perGralnick HR, Sultan C: The French-American-British (FAB) Cocentage may be seen long before clinical relapse (13). operative Group: Proposal for t h e classification of the myelodysQuantitation of nuclear DNA for the assessment of plastic syndromes. Br J Haematol 51:189-199, 1982. cell ploidy is still a n open question among cytometrists. 4. Cazzola M, Barosi G, Berzuini C, Dacco M, Orlandi E, Stefanelli M, Ascari E: Quantitative evaluation of erythropoietic activity in In our patients, the PI-DNA indexes were compared to dysmyelopoietic syndromes. Br J Haematol 50:55-62, 1982. those of bone marrow cells from control subjects. The 5. Crissman HA, Mullaney PF, Steinkamp JA: Methods and applinormal range obtained in this way was slightly wider cations of flow systems for analysis and sorting of mammalian than that obtained by Diamond et al. (6) in unfixed cells cells. Methods Cell Biol 9:179-246, 1975. 6. Diamond LW, Nathwani BN, Rappaport H: Flow cytometry in the stained with propidium iodide using chicken erythrodiagnosis and classification of malignant lymphoma and leukecytes a s a n internal standard. Comparative analysis of mia. Cancer 50:1122-1135, 1982. chromosome pattern and DI showed that a number of 7. Dosik GM, Barlogie B, Gohde W, Johnston D, Tekell JL, Drewinko patients with normal karyotype of bone marrow cells B: Flow cytometry of DNA content in human bone marrow: a had a slightly shifted DI with respect to the controls. critical reappraisal. Blood 55734-740, 1980. 8. Dresh C, Faille A, Poirer 0, Balitrand N, Najean Y: Bone marrow Abnormal DNA distributions had been occasionally seen cell kinetics and culture in chronic and subacute myelomonocytic in bone marrow cell populations without detectable leukemia. Physiopathological interpretation and prognostic imchromosome anomalies, due to the presence of aneuploid portance Leuk Res 4:129-142, 1979. clones escaping routine cytogenetic analysis (1,Z). How- 9 Dreyfus B: Preleukemic states. I. Definition and classification. 11. Refractory anemia with a n excess of myeloblasts in the bone ever, all patients with normal cytogenetic data and abA
A
I * s
-
*/
* . 4.
;ti. ::
v
Q
DNA FLOW CYTOMETRY IN MYELODYSPLASIAS marrow (smoldering acute leukemia). Blood cells 2:35-55, 1976. 10. Fried J: Method for quantitative evaluation of data from flow microfluorometry. Comput Biomed Res 9:263-276, 1976. 11. Hast R, Reizenstein P: Sideroblastic anemia and development of leukemia. Blut 42:203-207, 1981. 12. Hiddeman W, Buchner T, Andreef M, Wormann B, Melamed M, Clarkson BD: Cell kinetics in acute leukemia. A critical reevaluation based on new data. Cancer 50:250-258, 1982. 13. Hillen H, Burghouts J, Haanen C: The application of pulse cytophotometry for t h e early detection of relapse in acute leukemia. In: Third International Symposium on Pulse Cytophotometry, Lutz D (ed) European Press Medikon, Ghent, 1978, pp. 665-668. 14. KoeMer HP, Golde DW: Human preleukemia. Ann Intern Med 93:347-353, 1980. 15. Krishan A: Rapid flow cytofluorometric analysis of mammalian cell cycle by propidium iodide staining. J Cell Biol 66:188-193, 1975. 16. Mazzini G, Giordano P, Riccardi A, Montecucco CM: A flow cytometric study of human leukocytes and its relationship with chromatin structure. Cytometry 3:443-448, 1983. 17. Montecucco CM, Riccardi A, Traversi E, Mazzini G, Giordano P, Ascari E: Proliferative activity of bone marrow cells in primary dysmyelopoietic (preleukemic) syndromes. Cancer 1983 (in press). 18. Nicolini C, Kendall F, Baserga R, Dessaive C, Clarkson B, Fried J : The GdGl transition of W 138 cells. I. Laser flow microfluorometric studies. Exp Cell Res 106111-118, 1977. 19. Nowell P, Finam J: Chromosome studies in preleukemic states: IV. Myeloproliferative versus cytopenic disorders. Cancer 42:22542261, 1978.
243
20. Pet0 R, Pike MC, Armitage P, Breslow NE, Cox DR, Howard SV,
21.
22. 23.
24. 25.
26.
27.
Mentel N, McPherson K, Pet0 J , Smith PG: Design and analysis of randomized clinical trials requiring prolonged observation of each patient. 11. Analysis and examples. Br J Cancer 35:l-39, 1977. Riccardi A, Mazzini G, Montecucco CM, Cresci R, Traversi E, Berzuini C, Ascari E: Sequential vincristine, arabinosylcytosine and adriamycin in acute leultemia: cytologic and cytokinetic studies. Cytometry 3:104-109, 1982. Saarni MI, Linman JW:Preleukemia. The hematologic syndrome preceding leukemia. Am J Med 55:38-48, 1973. Seigneurin D, Hollard D: Use of tritiated thymidine-labelling index of the myeloblast-promyelocyte pool for the identification of the leukemic population in oligoblastic leukemia. Acta Haematol 66:181-186, 1981. Spitzer TR, Goldsmith J r GH: Myelodysplastic syndromes: is another classification necessary? Br J Haematol 52:343-344, 1982. Vindelev LL: Flow microfluorometric analysis of nuclear DNA in cells from solid tumors and cell suspensions. A new method for rapid isolation and staining of nuclei. Virchows Arch [Cell PatholI 24:227-242, 1977. Wickramasinge SN, Chalmers DG, Cooper EH: A study of ineffective erythropoiesis in sideroblastic anaemia and erythraemic myelosis. Cell Tissue Kinet 1:43-50, 1968. Zante J , Schumann J , Barlogie B, Gonde W, Buchner T: New preparation and staining procedures for specific and rapid analysis of DNA distributions. In: Pulse Cytophotometry 11, Schumann J, Gohde W, Buchner T (eds) European Press Medikon, Ghent, 1976, pp. 97-106.
