Subacute myeloid leukemia

Subacute Myeloid Leukemia A Clinical Review

JAMES R. WILLIAM PETER L. STANLEY Stunford,

COHEN, M.D.* P. CRECER. M.D. GREENBERG, M.D. L. SCHRIER. M.D.

California

From the Department of Medicine, Division of Hematology, Stanford University Medical Center, Stanford, California. This paper was supported in part by the Mildred Fertman Memorial Gift Fund, Division of Hematology, the Ricky Burton Memorial Gift Fund, Division of Hematology, and by Grant CA 05838.17, National Cancer Institute, National Institutes of Health, Bethesda, Maryland. Request for reprints should be addressed to Dr. Stanley L. Schrier, Department of Medicine, Division of Hematology, S-161, Stanford University Medical Center, Stanford. California 94305. Manuscript accepted January 17,1979. * Present address: Suite 202, 2410 Samaritan Drive, San Jose, California 95124.

The data on 31 patients who fit into the clinical spectrum of subacute myeloid leukemia have been reviewed. The majority of patients were male with a median age of 61 years. The interval from onset of symptoms to actuai diagnosis was extremeiy variable, with a mean of 16 months and a median of six months. Most patients presented with anemia and thrombocytopenia, although the white blood cell count varied from striking leukopenia to marked leukocytosis. Examination of the bone marrow invariably revealed abnormalities of all cell lines with megaloblastoid erythrogenesis and dysplastic megakaryocytopoiesis. Although the white cell line showed prominence of immature forms, there was more maturation than is seen in acute myeloid leukemia. Survival from diagnosis was variable, from less than one month to greater than 68 months, with a median of only six months. Anemia and hepatosplenomegaly were prognosticators of a poor outlook; patients with hepatosplenomegaly in association with either leukocytosis or thrombocytopenia had a particularly poor outlook, with a median survival of only one and a half months. Approximately half the patients received chemotherapy with no demonstrated effect on survival. The term subacute myeloid leukemia covers a clinical spectrum characterized by hypercellular bone marrow aspirates with abnormalities affecting all cell lines, often accompanied by peripheral cytopenias and circulating myeloblasts. The condition most commonly affects the middle-aged or elderly, most often males, with death occurring predominqntly from iflfection or bleeding. A variety of terms has been employed to describe this clinical picture, including subacute or chronic myelomonocytic leukemia [l-4], smoldering leukemia [5,6], chronic erythromonocytic leukemia or erythremic myelosis [‘T-IO], and refractory anemia with excess blasts [ll]. These terms refer to syndromes which appear to differ primarily in the degree of involvement of the erythroid versus myeloid cell line in the marrow, the degree of monocytosis and the duration of the illness. Although this disorder is described in the literature as “indolent” [2], most patients die within a year of diagnosis. There is little evidence in the literature to support chemotherapeutic intervention, with most investigators arguing that treatment is either not helpful [l] or indeed is contraindicted [I.& A few investigators, reporting on small series, do advocate chemotherapy [13,14], but most recommend only supportive care. Because of th6 lack of guidelines on this issue, a review of the cases of subacute myeloid leukemia seen at Stanford University Medical Center within the past several years was performed with particular attention being given to possible prognostic characteristics and to the effect of treatment on course and survival.

June 1979 The American Journal of Medicine

Volume 66

959

SUBACUTE MYELOID LEUKEMIA-COHEN ET AL

MATERIALS AND METHODS An attempt was made to collect the records of all patients with a diagnosis “subacute myeloid leukemia” at Stanford University Medical Center between 1970 and 1976. Because “subacute leukemia” is not included in the International Classification of Diseases (ICDA], some records may have been filed under other categories. Eventually 31 charts were obtained on which the data appeared to meet criteria for subacute myeloid leukemia. This number may well be an underestimate of the true prevalence of this disorder. The charts of these 31 patients were reviewed and, when applicable, letters and telephone calls were utilized to contact referring physicians, These 31 patients form the basis of this report. Information concerning 15 of these patients has appeared in a prior publication [15]. Thirty of the 31 patients are deceased, with information available about the terminal event in 28 of them. Bone marrow aspirates and clot sections were available for review in 28patients. The diagnosis of subacute myeloid leukemia in these patients had been made on clinical grounds by the hematologists caring for or consulting on them at the time of diagnosis at Stanford. All patients had abnormalities involving the granulocytic-monocytic series in peripheral blood or marrow but without clear-cut acute leukemia [vide infra). Because the clinical criteria utilized to make this diagnosis may have varied among the physicians, we have attempted to define characteristic morphologic features that would positively identify this group of patients as distinguished from those with myeloblastic or acute myelomonocytic leukemia and from those with chronic myeloid leukemia. The 28 marrow samples available were reviewed and then randomly mixed with 10 marrow samples from patients who had been considered on clinical and morphologic grounds to have AML. These slides were then reviewed blindly by two of us (JRC and WPC) and correlated with the initial diagnoses. In 90 per cent of the cases there was agreement on morphologic grounds with the clinical diagnosis. The most striking findings were as follows: (I] In subacute myeloid leukemia there was some maturation of the myeloid line beyond the promyelocyte stage, whereas in AML there was usually no maturation. (2) In subacute myeloid leukemia there were obvious abnormalities of all cell lines. Dysplastic and bilobed megakaryocytes were seen invariably. Minor degrees of megaloblastoid erythrogenesis were very common and in some Patients these changes were pronounced. (3) As noted from the spread in white blood cell count, peripheral smears were enormously variable. In 11 cases, myeloblasts were seen in the peripheral blood.

myelosis [7-10,161. These patients were looked at separately; because they did not differ significantly from the remaining 23 patients in any of the variables analyzed, they have been included within the entire group. RESULTS

Clinical Features. The patients’ ages ranged from 17 to 86 years with a mean of 60 years and a median of 61 years at the time of diagnosis. Twenty-two were male and nine female. Twenty-eight of the patients were Caucasian,, two Oriental and one black. The most common presenting signs and symptoms are shown in Table I. The most prevalent symptom complex prior to diagnosis included fatigue, shortness of breath and weakness, occurring in 74 per cent of the patients and most likely reflecting anemia. In addition, 49 per cent of the patients had some evidence of a bleeding diathesis. Other findings were less specific and included infection, headache, fever, sweats, anorexia, rash, dysphagia and recurrent phlebitis. The interval from the onset of symptoms to the time of diagnosis varied from less than a month to 109 months with a median of six months and a mean of 16 months. A history of possible myelotoxic exposure was identified in 14 patients, including prior therapeutic irradiation in two patients and pharmaceutical agents such as gold (one patient], chlorambucil (one patient) or anti-inflammatory agents (three patients). Six patients had exposure to industrial chemicals, and one patient had been a dentist using his own diagnostic x-ray equipment for many years [17]. Physical Examination. Physical findings at the time of diagnosis are outlined in Table II. When present, hepatomegaly and splenomegaly tended to be of modest degree. Lymphadenopathy occurred in just under a third of the patients. Petechiae or ecchymoses were found in 32 per cent of the patients on initial examination, correlating in degree with the platelet level. Five patients were found to have skin rashes: in two of these, a biopsy was performed and leukemic infiltration documented. Four patients (12.9 per cent) had gingival lesions and three (9.6 per cent] had retinal hemorrhage. Hematologic values at the Hematologic Parameters. time of diagnosis are shown in Table III. There was a wide range in the initial white blood cell count, but the majority were low with a median of 3,100/mm3. Similarly, the hemoglobin and hematocrit values ranged from quite low to normal but with low mean and median values of 9 g hemoglobin/d1 and 27 per cent hematocrit.

(4) Because monocytoid changes were previously claimed to be useful [1,3-41, we analyzed our patients and found that these features were present in peripheral blood or bone marrow in 50per cent of the patients with either AML or subacute myeloid leukemia and hence did not help to differentiate the disorders.

The platelet counts ranged from 4,000 to 334,000/mm3 with a mean of 91,000/mm3 and a median of 56,000/ mm3. The peripheral blood smears were remarkable for a variety of red cell morphologic changes (Table IV), most commonly macro- and microcytosis with microangiopathic changes, nucleated red blood cells,

In eight of the 31 patients, red blood cell and bone marrow morphologic changes were so predominant as to lead to a clinical diagnosis of DiGuglielmo’s disease or erythremic

teardrop forms and polychromatophilia. Less frequent findings included hypochromia, basophilic stippling, rouleaux, target cells, burr cells and spherocytes. Cells

960

June 1979 The American Journal of Medicine Volume 66

SUBACUTE

MYELOID LEUKEMIA-COHEN

ET AL

of the granulocyte-monocyte series often appeared dysplastic, occasionally with Dbhle bodies, toxic granulation and pseudo-Pelger-Huet phenomenon. Monocytosis was common but not universal. The platelets often appeared large and hypogranulated. Examinations of bone marrow aspirates stained with Wright-Giemsa and marrow clot sections stained with hematoxylin and eosin were reviewed in 28 of the 31 patients. In the majority of cases, the over-all cellularity

was increased, but in five the cellularity was normal and in five others the marrow appeared hypocellular [Table V], In all cases the myeloid series appeared to contain numbers of immature forms, but there was some maturation to more mature forms. Myeloblasts tended to occur in clumps as seen on spicule section, and no attempt was made to obtain differential counts. Over-all appearance was thought to be more useful. In a number of cases the promyelocytic granules appeared strikingly

TABLE I

TABLE II

Presenting Signs and Symptoms SignorSymptom

Weakness or fatigue Shortness of breath Bleeding Bruising or petechiae Weight loss Infection Fever Dizziness or confusion Headache Anorexia Recurrent phlebitis Other

NO.

%

14 9 a 7 5 5 3 3 3 2 2 7

45 29 26 23 16 16 10 10 10 6 6 23’

Physical Findings at Diagnosis

Hepatomegaly Splenomegaly Petechiaelecchymoses Lymphadenopathy Hepatosplenomegaly Skin rash Gingival lesions Retinal hemorrhage

TABLE IV

-

No.

%

12 12 10 9 a 5 4 3

38.7 38.7 32.0 29.0 26.0 16.0 12.9 9.6

Red Blood Cell Abnormalttles

One each with bone pain, sweats, dysphagia, arthralgias, subcutaneous nodules, rash, and palpitations. l

P8tkttS

Finding

TABLE III

Macrocytosis Microcytosis Microanglopathic changes Nucleated red blood cells Hypochromia Teardrop forms Polychromatophilia Other’

Hematologic Parameters at Diagnosis Mean

Median

0.8-240

16.5

3.1

4.6-13.4 14.5415 4-334

9.0 27.3 91.0

9.2 27.2 56.0

Parameter

Ran98

White blood cells (thousand/mm3) Hemoglobin (g/dl) Hematocrit (%) Platelet (thousand/

No.

%

19 19 14 12 9 9 a 7

61 61 45 39 29 29 26 23

Includes basophilic stippling, rouleaux, targets, burrs and spherocytes. l

mm3)

TABLE V

Bone Marrow Morphology

-

Parameter Over-all cellularity Iron pattern Erythroid cellularity Megakaryocyte number

TABLE VI

Increased

Decreased

Normal

Unknown

No.

%

No.

%

No.

%

No.

ia 13 11 4

64 46 39 14

5 7 17 17

ia 25 61 61

5 6

ia 21

2

%

. 7

. 7

25

Muramidase and Leukocyte Alkaline Phosphatase

Parameter Muramidase Leukocyte alkaline phosphatase

Normal Range

Patients (total no.)

15. 48 50-100

28 18

increased No. % 14 4

50 22

Normal No. % a 4

29 22

D8cre888d No. % 6 10

21

56

NOTE: Muramidase units; equivalents of crystalline egg white muramidase in micrograms per milliliter; leukocyte alkaline phosphatase units: an arbitrary score derived by counting 100 polymorphonuclear leukocytes and grading the intensity of staining from 0 to 4.

June 1979

The American

Journal

of Medicine

Volume 66

961

SUBACUTE MYELOID LEUKEMIA--COHEN

ET AL

coarse and condensed. Auer rods were seen in only four patients (14per cent]. The erythroid series, although varying in cellularity, was in all cases strikingly megaloblastoid. Ringed sideroblasts were seen in only eight patients (29 per cent). No consistent iron pattern was apparent. Megakaryocytes were thought to be decreased in most but not all marrows, but in virtually all cases they appeared dysplastic with mono and bilobed nuclei. Leukocyte alkaline phosphatase and serum muramidase values are shown in Tabl’e VI. No consistent abnormalities in serum vitamin Blz or folic acid were seen. Chromosomal analyses were performed in too few patients for comment. Clinical Course. Survival from diagnosis ranged from less than one month to 68 months in the 28 patients with known date of death. One patient is still alive two and a half years from diagnosis. The mean survival from diagnosis is 10.3months with a median of six months (Figure 11. Survival data were examined with regard to sex, white blood cell count, hematocrit value, platelet count, presence of adenopathy, leukocyte alkaline phosphatase, serum muramidase, presence of organomegaly and marrow over-all, and erythroid cellularity. Statistically significant differences were found only in those patients who presented with anemia, who had a median survival of 5.7 months as contrasted with 15.2 months for nonanemic patients [p
962

June 1979

The American Journal of Medicine

pneumonia documented although two patients had unexplained pulmonary infiltrates with nondiagnostic lung biopsies. An attempt was made to assess the impact of treatment on survival. Thirteen patients were treated with a variety of regimens. Ten received combinations containing daunomycin. Nine of these also received cytosine arabinoside and 6-thioguanine (DAT), and one patient was treated with daunomycin and prednisone (DP). Other regimens included combinations of cyclophosphamide, vincristine, cytosine arabinoside and prednisone (COAP) or various combinations without one or more of these agents (Table XII). Several patients received more than one regimen during the course of their illness. The mean interval from the time of diagnosis to the onset of treatment was three and a half months with a range from 0 to 17 months and a median survival after the initiation of treatment of only one and a half months. There was no significant difference in the over-all survival of the treated patients as compared with the untreated patients (Table XIII, Figure 4). COMMENTS Whereas the clinical and morphologic features of both acute and chronic myeloid leukemia are well known and widely accepted, subacute myeloid leukemia is a less well-defined clinical entity. Much of the literature on subacute myeloid leukemia is concerned with the ultrastructure [18-201, cytochemistry [21,22] or nature of the malignant cell of origin [l-4,7-10]. The definition of subacute myeloid leukemia was based on marrow morphology and clinical presentation. It is of interest that, on blind reading of peripheral blood and marrow smears, monocytosis was not a helpful differential point in distinguishing subacute from acute myeloblastic or myelomonocytic leukemia. Although abnormalities were present in all cell lines, eight of the patients in this series had been thought clinically to have a DiGuglielmo variant of leukemia based on the predominance of abnormal erythroid hypercellularity in the marrow specimen [23]. It .is of interest that these patients did not differ from the other patients in this series in any of the parameters examined, including presenting hematocrit values, white blood cell and platelet counts, degree of marrow cellularity and response to treatment or survival. There is most likely some overlap between these patients with subacute myeloid leukemia and those described in the literature as “preleukemic” [24-261. Both groups of patients often present with prolonged cytopenias and dysplastic marrow morphology. Unlike preleukemia, however, patients with subacute myeloid leukemia can clearly be called leukemic at the time of diagnosis, based on the presence of increased myeloblasts in the marrow with or without circulating myeloblasts as well. The clinical features of the patients presented here are similar to those described in the literature. This ill-

Volume 66

SUBACUTE

MYELOID

LEUKEMIA-COHEN

ET AL

ACTURRIAL SURVIVAL

:

ENTIRE POPULFITIBN I I I

I

I

I 10

I 20

I

I-

I 60

II 70

60-

1 : a

40-

m Q LT Q. M-

, OO

Figure 1.

l

I

TI;E

I

I

[ f14~NTH~l

Entire population-actuarial

90

survival

FiCTUflRIRLSURVIVAL

HEMRTOCRITI I

I

I

LOW VS. NORMAL I

I

I

I

Yaw73

L_______-, , 11 0, tI

OO

1 10

I 20

: _‘_____C__________________________________~ 1 I I 1 I I I I I I 80 m

T I ;E

I f14; N T H toI

Figure 2. -,

TABLE VII

LOW(<35

%)

Normal (>35 %)

J

Actuarial survival versus anemia: - - -, hernatocrit value >35 per cent; hematocrit value <35 per cent.

Survival Versus Hematocrit

HWl!&C?ii

80

(p <0.04)

Patients

Mean

(no.)

(mo)

(W

23

8.5

5

25.6

5.7 15.2

Median

ness most commonly affects the middle-aged or elderly, usually males, with the diagnosis becoming apparent after a variable but often prolonged symptomatic course. Death, as in acute myeloblastic leukemia, is most often due to infection or hemorrhage. The finding of a median survival from diagnosis of only six months suggests that, once clinically apparent, subacute myeloid leukemia

June 1979

The American Journal of Medicine

Volume 66

963

SUBACUTE MYELOID LEUKEMIA-COHEN

ET AL

Figure 3. Influence of hepatosplenomegaly hepatosplenomegaly; - - -, megaly; -, tosis and thrombocytopenia.

TABLE IX

Survival Versus Hepatomegaly or Splenomegaly

TABLE VIII

PhysicalFinding No hepatomegaly or splenomegaly Hepatomegaly only Splenomegaly only Hepatosplenomegaly

Mean (mo)

Median (mo)

14

15.0

10.9

4 3 8

11.7 10.5 5.6

8.2 9.8 2.9

Patients (no.1

on survival: - - -, no hepatosplenohepatosplenomegaly with leukocy-

Age(vr)

Patients (no.1

Mean (mo)

Medlan (mo)

150 51-60 61-70 >70

5 9 12 5

15 8 6 20

17 6 4 15

TABLE XI TABLE X

Localized Infections Leading to Systemic Illness

Primary Cause of Death (21 patients)

Infection Bacterial Fungal Bleeding Gastrointestinal Cranial Acute myocardial infarction Renal failure Blastic chanae

No.

%

11 8 3 6 5 1 2 1 1

52.5

lnfsction Fungus Pneumonia Cellulitis Upper respiratory tract infection Perirectal abscess Unexplained pulmonary infiltrate Meningitis Pyelonephritis Deep abscess

G.b

‘9.5 5.0 5.0

is a grave illness. Over a similar time period, patients with acute myeloblastic leukemia treated at Stanford University Medical Center have had a median survival in excess of 15 months [27]. The lack of impact of treatment on survival is discouraging, although a systematic analysis of treatment 964

Survival Versus Age

June 1979 The American Journal of Medicine

Episodes No. % 6 3 3 3 2 2 1 1 1

27.0 14.0 14.0 14.0 9.0 9.0 4.5 4.5 4.5

modalities and response rate is not possible from this kind of analysis. In comparing our treated group of patients with those left untreated, there was no significant difference in any of the signs or symptoms just discussed. Although there tended to be more patients with hepatosplenomegaly in the treated group than in the

Volume 66

SUBACUTE MYELOID LEUKEMIA-COHEN

ET AL

ACTUARIRL SURVIVAL

UJO-

I

I

3t3nm_

I I

Y g

TREATED VS. UNTREATED I I I I

I

40-

m 2 L 2u-

OO

Figure 4.

1 10

Actuarial

IL- ---__-, 8 L_______, I L_-__-__-_-____________________r-__-~ I I 1 I I I 80 m 20

T I;E

survival-influence

I P14iN TH-?I

of treatment:

- - -, untreated;

80 -,

treated.

TABLE XII

Treatment

Daunomycin, Ara-C (cytosine arabinoside), thioguanine Cyclophosphamide, Oncovin (vincristine), Ara-C, prednisone Prednisone Cyclophosphamide, vincristine, prednisone Vincristine, prednisone Daunomycin, prednisone l

TABLE XIII

Treatments Employed Patients’ ho.1

Treated Untreated

9 3 3 2 2 1

Some patients received more than one regimen.

untreated group, this difference did not achieve statistical significance. The reasons for choosing chemotherapy in a given patient varied and represented the clinical decision of the physician caring for the patient at the time, based usually on an accelerating white blood cell count, refractory thrombocytopenia or frequent, significant infectious problems. For those patients treated most aggressively with daunomycin-containing regimens, the median survival from the onset of treatment was only one and a half months. Whereas this undoubtedly reflects to some extent the critical condition of some of these patients, it also suggests that this might not be the optimum treatment for this stage of this illness. It is possible that we are waiting too long to initiate chemotherapy, that we should be treating the patient earlier before our hands are forced by infection or bleeding. Physicians are loath to do this in a setting in which they believe some patients may have a relatively long prognosis if left untreated, and in which the potential morbidity and mortality of aggressive chemo-

Survival Versus Treatment (p = 0.4) Patients (no.)

Mean (mo)

Median W)

13 15

8.1 14.6

5.6 11.2

therapy are, indeed, significant. Because of this, we attempted to find a means of distinguishing those patients at greatest risk of short survival. In our statistical analysis, only hepatosplenomegaly and anemia achieved statistical significance as prognosticators of potentially short survival. When hepatosplenomegaly was accompanied by either thrombocytopenia or leukocytosis, there was a particularly poor outlook with a median survival in this group of only one and a half months, although the number of patients was too small for statistical significance. In an earlier study of patients with a variety of myeloproliferative syndromes, Greenberg et al. [l5] showed that changes in in vitro colony-forming capacity and an increase in abnormally light density granulocyte precursors did occur in patients with subacute myeloid leukemia and presaged clinical deterioration. It is possible that utilization of in vitro techniques, combined with an assessment of the risk factors discussed, would help to select those patients in whom chemotherapy should be initiated. A systematic prospective study is necessary to evaluate the treatment regimen of choice for this disease. ACKNOWLEDGMENT We wish to acknowledge Ms. Gail Thompson for assistance with the preparation of the manuscript.

June 1979 The American Journal of Medicine

Volume 66

965

SUBACUTE MYELOID LEUKEMIA-COHEN

ET AL

REFERENCES 1. Zittoun R Subacute and chronic mvelomonocvtic leukaemia: a distinct haematologic entity. Br ] Haemaiol32: 1,1976. 2. Sexauer J, Kass L, Schnitzer B: Subacute myelomonocytic leukemia. Am J Med 57: 853,1974. 3. Miescher PA, Farquet JJ: Chronic myelomonocytic leukemia in adults. Semin Hematol 11:129,1974. 4. Geary CG, Catovsky D, Wiltshaw E, et al.: Chronic myelomonocytic leukeamia. Br J Heamatol30: 289,1975. 5. Rheingold JJ, Kaufman R, Adelson E, et al.: Smoldering acute leukemia. N EnelI Med 268: 812.1963. 6. Knospe WH, Greg&$ SA: Smoldering acuteleukemia. Arch Intern Med 127: 910,197l. 7. Broun GO: Chronic erythromonocytic leukemia. Am J Med 47: 785, 1969. a. Shaw MT, Bottomlev SS, Bottomley RH, et al.: The relationship of erythromonocytic leukemia to other myelooroliferatiie disorders. Am 1Med 55: 542.1973. 9. Schwartz SO, Critchlow J: Erythremic myelosis (DiGuglielmo’s disease). Blood 7: 765,1952. 10. Kass L: Chronic erythremic myelosis. Postgrad Med 58: 161, 1975. 11. Bennett JM, Catovsky D, Daniel M-T, et al.: Proposals for the classification of the acute leukaemias. Br J Haematol 33: 451,1976. 12. Crosby WH: To treat or not to treat acute granulocytic leukemia. Arch Intern Med 122: 79.1968. 13. Bloomfield CD, Brunning RD: Prognostic implications of cytology in acute leukemia in adults. The case for subacute leukemia. Hum Path01 5: 641.1974. 13. Bloomfield CD, Brunning RD: Prognostic implications of cytology in acute leukemia in adults. The case for subacute leukemia. Hum Path01 5: 641,1974. 14. Bloomfield CD, Brunning RD. Kennedy BJ: DaunorubicinPrednisone treatment of erythroleukemia. Ann Intern Med

966

June 1979

The American Journal of Medicine

15. 16. 17. 18. 19.

20.

21. 22. 23.

24. 25. 26. 27.

Volume 66

81: 746, 1974. Greenberg P, Mara B, Bax I, et al.: The myeloproliferative disorders. Am J Med 61: 878,1976. Dameshek W: The DiGuglielmo syndrome revisited. Blood 34: 567,1969. Potolsky A, Creger WP: Radiation and drug therapies and leukemia. Ann Rev Med 24: 75,1973. Freeman AI, Journey LJ: Ultrastructural studies on monocytic leukemia. Br J Haematol20: 225,1971. Glick AD, Horn RG: Identification of promonocytes and monocytoid precursors in acute leukaemia of adults: ultrastructural and cvtochemical observations Br 1Haematol 26: 395,1974. ” Skinnider LF, Card RT, Padmanabh S: Chronic myelomonocytic leukemia. Am J Clin Path01 67: 339,1977. Hayhoe FGJ, Cawley JC: Acute leukemia: cellular morphologv, cvtochemistrv and fine structure. Clin Hematol i: 49, %72: Shaw MT: The cvtochemistrv of acute leukemia: a diagnostic and prognostic evaluation. Semin Oncol3: 219,19?6. Eastman PM, Schwartz R, Schrier SL: Distinction between idiopathic ineffective erythropoiesis and DiGuglielmos’s disease: clinical and biochemical differences, Blood 46: 487,1972. Saarni MI, Linman ]W: Preleukemia. The hematological syndrome preceding acute leukemia. Am J Med 55: 36, 1973. Linman 1W: Mvelomonocvtic leukemia and its nreleukemic phase: J Chronic Dis 22: 713,1979. Saarni MI, Linman JW: Myelomonocytic leukemia: disorderly proliferation of all marrow cells. Cancer 27: 1221, 1971. Embury SE, Elias L, Heller PH, et al.: Remisson maintenance theraov in acute mveloeenous leukemia. West 1Med 126: