(1998) 12, 1071–1075  1998 Stockton Press All rights reserved 0887-6924/98 $12.00 http://www.stockton-press.co.uk/leu Acute myeloblastic leukemia with minimal myeloid differentiation: phenotypical and ultrastructural characteristics N Villamor1, M-A Zarco2, M Rozman1, J-M Ribera2, E Feliu2 and E Montserrat1

1Servei d’Hematologia, Hospital Clinic de Barcelona; and 2Servei d’Hematologia, Hospital Universitari Germans Trias i Pujol, Badalona, Spain

AML-M0 is an infrequent form of acute myeloblastic leukemia Material and methods characterized by negative reaction with myeloperoxidase (MPO), Sudan Black and lymphoid antigens and positivity for CD13 or CD33. In the present study we describe the immuno- Patients phenotypical and ultrastructural characteristics of a group of AML-M0 in adult patients. Nine out 218 AML (4.1%) During a 4 year period 82 de novo acute lymphoblastic leukem- fulfilled the AML-M0 criteria. CD13 or CD33 were positive in ias, 218 de novo acute myeloblastic leukemias and five undiffer- eight out nine cases, with two or more positive myeloid anti- entiated leukemias were diagnosed in the Hospital Clinic de gens being present in 82% of the cases. Immunological MPO Barcelona, Barcelona, and the Hospital Universitari Germans was positive in 57% of the cases and CD68 in 33%. In no case were megakaryocytic and erythroid markers present. Four Trias i Pujol, Badalona, in patients older than 15 years. cases (44%) expressed CD7 and TdT but only two coexpressed both antigens. In none of the cases was CD3 or CD22 cytoplas- mic expression found. Ultrastructurally, a low number of gran- Classification of acute leukemias ules was seen in all cases whereas ferritin particles or rhopheo- cytosis were not observed. Ultrastructural MPO was positive in Morphology and cytochemistry: Acute leukemias were one out of five cases and platelet peroxidase (PPO) was nega- classified according to the FAB criteria.1–4 Peripheral blood tive in the four cases studied. Two out of six cases showed ¨ karyotypic abnormalities (hypotetraploidy and a complex kary- and smears were stained with May–Grunwald– otype, respectively). In two out three cases a rearranged pat- Giemsa. Cytochemical reactivity against myeloperoxidase ␤ ␥ tern for JH gene was observed. TCR (C and J ) rearrange- (MPO), Sudan Black, naphtol-ASD-acetate with FINa inhi- ments were not detected in any case. AML-M0 is an infrequent bition and alpha-naphtyl acetate was analyzed. form of acute myeloblastic leukemia. A large panel of myeloid monoclonal antibodies (MoAb) and the study of the cytoplas- mic expression of myeloid antigens is necessary to diagnose this form of leukemia. AML-M0 usually coexpress lymphoid Immunophenotype: Immunocytochemical analysis by markers. Ultrastructural studies may be of help to discard an alkaline phosphatase anti-alkaline phosphatase technique5 immature erythroid proliferation. and/or immunofluorescence technique and flow cytometry of Keywords: ; AML-MO; ultrastructure; peripheral and/or bone marrow samples were performed to immunophenotype determine the expression of lymphoid and myeloid antigens in blast cells. Reactivity against B and T lymphoid antigens (Tdt, CD19 or cytoplasmic CD22, CD10, CD7, cytoplasmic Introduction CD3) as well as myeloid antigens (CD33 or cytoplasmic CD13, CD14, CD68, myeloperoxidase) and HLA-DR The French–American–British (FAB) co-operative group pro- expression was investigated. In addition, in most cases the posed morphological, cytochemical and immunophenotyp- presence of CD2, CD5, CD11b, CD41 or CD61, CD36, glyco- ical criteria to identify a new type of acute myeloblastic leuke- phorin A, and CD34 was tested. An acute leukemia was con- mia (AML) with minimal myeloid differentiation which is sidered as positive for an antigen when more than 20% of the known as AML-M0.1 This is an infrequent form of leukemia blast cells reacted with it (Ͼ10% in the case of immunological which, according to the FAB group, represents less than 5% myeloperoxidase (iMPO). of all acute myeloblastic leukemias. The main characteristic of this new AML subtype is the agranular cytoplasm of the blast cells resembling lymphoblastic cells and the negative Ultrastructural morphology: Mononuclear cells of periph- cytochemical reactions for myeloperoxidase and Sudan Black. eral blood or bone marrow were processed by electron The myeloid origin of the blast cells is demonstrated by the microscopy. Cells were fixed in 1.25% glutaraldehyde in 0.1 M presence of myeloid antigens (CD13 or CD33) and the lack cacodylate buffer, postfixed in fresh 1.25% glutaraldehyde for ° of expression of lymphoid markers. 2h at4C, followed by 1% OsO4 and dehydrated through a We describe the characteristics of a series of nine patients graded series of ethanol. Dehydrated samples were included in with AML-M0 with special emphasis on the immunopheno- Durcupan. Thin sections were cut on a Reichert’s ultramicro- typic pattern and the ultrastructural features of these cases. tome, mounted on copper grids, stained with uranyl acetate and lead citrate and examined in a Philips 410 electron micro- scope.6 In each case a minimum of 25 blast cells was studied, counting the number of granulated blast cells, the number of granules per blast section, the nuclear outline and the presence of cytoplasmic organelles. To detect ultrastructural myeloperox- Correspondence: Prof E Montserrat, Servei d’Hematologia, Hospital Clinic de Barcelona, c/Villarroel 170, 08036 Barcelona, Spain; Fax: idase bone marrow particles or buffy-coat cells were fixed for 343 227 54 75 30 min at 4°C with 1.25% distilled glutaraldehyde in 0.1 M Received 25 September 1997; accepted 24 March 1998 phosphate buffer, pH 7.2. Cells were washed three times in the Phenotypic and ultrastructural characteristics of AML-MO N Villamor et al 1072 buffer. Samples were incubated for 30 min at room temperature Results in the dark in DAB medium. The cells were then rinsed in sev- eral changes of phosphate buffer, 0.1 M pH 7.2, and postfixed The FAB distribution of AML was as follows: nine M0, 36 M1, for 30 min in osmium tetroxide in phosphate buffer. After post- 30 M2, 37 M3, 30 M4, 64 M5, nine M6, three M7. Thus, the fixation the cells were dehydrated in graded ethanol solutions frequency of AML-M0 in our AML series was 4.1%, without and embedded in epoxyresin. Ultrathin sections were examined differences between the two centers contributing cases. The either unstained or slightly stained with lead citrate.6 Ultra- main clinical and analytical data of these patients as well as structural platelet peroxidase: cells were fixed in 1.25% glutaral- their outcome are depicted in Table 1. dehyde in Gey’s buffer at 4°C for 10 min, washed three times with the same buffer and incubated in DAB medium for 60 min at room temperature in the dark; the remaining steps were car- Morphological and cytochemical characteristics ried out as in usual MPO staining.6 The morphological appearance of blast cells was lymphoid in seven cases (78%) and myeloid in two (22%). The myelo- Karyotypic analysis peroxidase and Sudan Black reactions were negative in all cases, while esterases were negative in all cases but two which showed low intensity reaction in a few blast cells. In Chromosomal analysis was performed in peripheral or bone the three patients who relapsed (Nos 1, 4, 7) after having achi- marrow specimens. Mononuclear cells were cultured in RPMI eved a remission the morphological and cytochemical charac- 1640, supplemented with 20% heat-inactivated fetal calf ␮ teristics of the blast cells at relapse were the same as those at serum enriched with 2% L-glutamine, 100 g/ml penicillin, diagnosis. In one of the two patients with refractory disease ␮ 10 g/ml streptomycin, 1% sodium heparin. Chromosome (No. 2) the appearance of Auer rods in 15% of cells and mye- preparations were obtained after 24–48 h of incubation. The loperoxidase restricted to the Auer rods was noticed 6 months slides were stained with a G-banding. Well-spread meta- after the diagnosis. phases were photographed and then arranged according to the recommendations of the International System for Human 7 Cytogenetic Nomenclature. Immunophenotype

Table 2 shows the immunophenotype of AML-0 cases. All cases Organization of immunoglobulin gene and beta chain were negative for B lineage antigens (CD19, cytoplasmic CD22 gene of T cell receptor and CD10). Cytoplasmic CD3 was negative in all cases, one case was CD2 positive (11%) and four cases were CD7 positive + High molecular weight DNA was obtained by lysis of leu- (44%). The mean percentage of CD7 blast cells was 59% (s.d. kemic cell samples with sodium dodecylsulphate (SDS) and 27, range 35–91%). Four cases (44%) were Tdt positive with a proteinase K, followed by extraction with phenol-chloroform mean value of 43% (s.d. 21, range 20–63%). Only two cases and chloroform and further precipitation with ethanol.8 The coexpressed CD7 and TdT. HLA-DR was positive in all but one DNA samples were digested with restriction endonucleases case and CD34 in five out of eight cases (63%). Six patients out EcoRI, and HindIII (Boehringer Mannheim, Mannheim, of nine (67%) were CD13 positive with a mean value of 69% Germany). DNA fragments were size fractioned by electro- (s.d. 18, range 44–95%). Four out of six cases (67%) were CD33 phoresis in 0.8% agarose gel and transferred to nitrocellulose positive with a mean value of positive cells of 47% (s.d. 10, filters. The probes were labeled by a random priming method range 35–58%). Three out of nine cases (33%) were CD68 posi- to a specific activity of 1 to 2 × 109 c.p.m./␮g of DNA. After tive with a mean positivity of 62% (s.d. 19, range 51–84%). hybridization, the filters were washed stringently at 65°C, CD14 was positive in one out of seven cases (14%) and CD11b dried, and autoradiographed. Ig gene analysis was performed was positive in two out of seven cases (29%). In seven AML- using a 2.5 Kb EcoRI/PvuI fragment corresponding to the join- M0 the expression of myeloperoxidase was analyzed by immunological technique (iMPO). In four cases, more than 10% ing region of the immunoglobulin heavy chain (JH). The organization of T cell receptor was analyzed using a 0.8 kb of the blast cells were positive with a mean percentage of 27 (HindIII/EcoRI) fragment for the joining region of the T cell (s.d. 13, range: 11–42%). In most cases the blast cells reacted receptor ␥ chain gene (J␥) and a 0.6 kb (HindIII/EcoRI) frag- with more than one myeloid antigen: two cases (22%) expressed one myeloid marker (CD13 and CD33); three (33%) were posi- ment for constant region of the T cell receptor ␤-chain (C␤). The presence of breakpoints in the major breakpoint cluster tive for two myeloid antigens (CD13 and CD14, CD68 and region was investigated using bcr-3 probe (MBR) with a 2 kb iMPO, CD13 and CD68) and four (44%) reacted with three fragment corresponding to the M-bcr gene. These probes were myeloid antigens (CD13, CD33 and iMPO; CD13, CD33 and kindly provided by Dr L Luzzatto, from the Department of CD11b; CD33, CD11b and iMPO; CD13, CD68 and iMPO). Human Genetics, Memorial Sloane Kettering Cancer Center, No case was positive for megakaryocytic antigens (CD41 or New York, USA. CD61) or glycophorin A. Table 3 compares the immunopheno- type of AML-M0 with the other AML subtypes. CD7 was found to be more frequently expressed in AML-M0 than in the other AML varieties. In contrast, iMPO expression was less frequent Statistical analysis in AML-M0 than in the other subtypes. In three patients the immunophenotype was also analyzed at Quantitative values were expressed as mean ± standard devi- relapse. In patient No. 4 the percentage of blast cells expressing ation (s.d.). Student’s t-test was used for comparing quantitat- CD33 increased and iMPO became positive, and in patient No. ive values between AML groups and the ANOVA test for 7 the percentage of CD33-positive blast cells increased whereas qualitative parameters. CD14 became positive and CD7 expression was lost. Phenotypic and ultrastructural characteristics of AML-MO N Villamor et al 1073 Table 1 Clinical and analytical characteristics of AML-M0 patients

Patient No. 123456789

Age/sex 56/M 61/M 55/M 23/M 71/F 72/F 39/M 80/M 68/F Adenomegaly No No No No No No Yes Yes No Hepatomegaly No No No No No No No No 2 cm Splenomegaly No No No No 3 cm No No No No Peripheral blood Hemoglobin (g/l) 63 115 56 125 68 102 122 90 80 Leukocytes (×109/l) 2.8 1.7 6.7 18.0 1.2 1.4 115 155 1.8 Blasts cells (%) 22 50 56 92 8 0 79 98 22 Platelets (×109/l) 128 233 * 66 84 202 204 29 60 Bone marrow Morphology My Ly Ly Ly Ly My Ly Ly Ly Blasts cells (%) 58 48 70 99 51 71 83 58 47 Myeloperoxidase 1.4 1.3 1.6 2.4 0 0002 Sudan Black 000000000 ANAE ±−30 −−±−−− Treatment Chemotherapy DAE DAE DAE DAE1 No DAE DAE2 MX-ARA DAE Complete remission Yes No Yes Yes * No Yes * Yes Relapse (months) Yes (48) * No Yes (15) * * Yes (10) * No Survival3 53 27 6+ 21 3 28 20 0.5 63+

M, male; F, female; My, myeloid; Ly, lymphoid; *, not done or not applicable; −, negative. Chemotherapy (administered according to Sierra et al)29: DAE, daunorubin, cytarabine and etoposide; 1followed by ABMT, 2ABMT in second remission; survival3: + alive.

Table 2 Immunophenotypical results of AML-M0 patients

Patient 123456789

TdT 32 −− 20 −− − 58 −− − 63 DR 79 85 78 67 78 12 18 65 98 85 62 99 CD34 44 74 7 78 7 * 52 * 77 94 86 9 CD19 −−−−−−−−−−−− cCD22 * −− −−−−−−−−− CD10 −−−−−−−−−−−− cCD3 −−−−−−−−−−−− CD7 35 35 − 97294− 39 91 −− − CD2 −− − − −− − * −− 34 * CD5 −− − − −− − * −− − − cCD13 68 83 95 54 −− 75 − 75 78 4 44 CD33 58 58 * − 35 66 46 * 10 97 50 * CD14 −− 40 −−−− * − 20 − * CD11b 17 12 * −−620 −−14 61 * CD68 − 3 −−−13 2 52 84 * 16 51 CD36 * −− **− * − * −− * CD41/CD61 −−−−−−−−−−−− MPO 42 7 * −−8* 30− 12711 Glyco-A −− * −−** *− ** * Methodology C C C I C C C I C C C 1

*, not done; −, negative. Second column, results at relapse. C, combined flow cytometry and immunocytochemistry. I, immunocytochemistry.

Cytogenetics and molecular biology patients studied. In contrast, two patients showed a rearranged

pattern in the JH region of the immunoglobulin gene (see Table A sufficient number of metaphases was obtained in six cases 4). None of the patients showed rearrangements in the bcr gene. with four normal karyotypes, one case with hypotetraploidy (79-88; XXXX [20]), and one case with complex karyotype with 13 metaphases; in spite of bad morphology preventing a Ultrastructure clear definition of the breakpoints it was possible to identify the following abnormalities [46XY, t(6;11)(q?;q23)/46XY Ultrastructural analysis was performed in eight patients. In all del(6)(q?)/46XY del(3) (q2)]. cases, the blast cells showed cytoplasmic granulation (ranging TCR ␤-chain and J␥ configuration was normal in the three from 4 to 80% of blasts). The number of granules per section Phenotypic and ultrastructural characteristics of AML-MO N Villamor et al 1074 Table 3 Immunological characteristics of AML-M0 compared with terion for the diagnosis of this subtype of AML. Most of our a the other FAB subtypes cases (78%) were positive for two or more myeloid antigens. The high heterogeneity in the antigens expressed is of note. M0 M1–M2 M3 M4–M5 P Interestingly, only the use of CD13 and CD33 may be insuf- 10–12 + ficient for the diagnosis of AML-M0. Thus, the use of a CD7 3/6 6/38 0/14 7/47 0.02 large panel of myeloid MoAb and the analysis of cytoplasmic CD7(%)b 35.0 ± 39.9 5.7 ± 17.3 0 ± 0 2.7 ± 7.4 expression of the antigens is often necessary to detect these CD68+ 1/6 17/37 11/11 32/40 0.001 cases.13 In half of our cases immunological MPO was positive. CD68(%)c 11.2 ± 20.1 28.1 ± 33.7 75.1 ± 22.8 40.6 ± 30.1 The discordance between cytochemical and immunopheno- iMPO+ 1/5 30/34 14/14 20/38 0.001 typical MPO expression has been previously referred to, this iMPO(%)d 12.1 ± 17.7 64.3 ± 39.7 97.4 ± 7.0 27.4 ± 34.7 being particularly frequent in the poorly differentiated AML 10,12,14 CD14 1/6 6/35 2/13 15/38 NS subtypes. Although in the FAB criteria the negativity of CD14(%) 8.0 ± 17.9 11.0 ± 23.4 1.8 ± 5.7 24.1 ± 29.7 lymphoid markers was considered as a criterion for the diag- nosis of AML-M0, a high percentage of such forms of leukemia TdT+ 2/6 6/38 0/14 7/47 NS TdT(%)e 19.0 ± 25.3 9.4 ± 21.1 0.1 ± 0.3 9.6 ± 22.0 express lymphoid antigens, mainly CD7 and TdT, as observed in our series and others.9,11,15–18 The minimal myeloid features aData from the Department of Hematology, Hospital Clinic, Barcelona. and the high frequency of CD7 or TdT expression may pose bP Ͻ 0.05 between M0 and the other three subgroups. diagnostic difficulties. In fact, acute leukemias sharing ident- cP Ͻ 0.05 between M3 and the other three subgroups. ical immunophenotypic patterns can be differently classified dP Ͻ 0.05 between M0 and M3, M0 and M1–M2, M4–M5 and M3, (eg biphenotypic leukemias included in the AML-M0 subtype and M4–M5 and M1–M2. or the opposite) depending on the studies.19,20,21 In order to e Ͻ P 0.05 between M0 and M3. overcome this problem, strict criteria to diagnose bipheno- NS, not significant. typic leukemias have been proposed.22 None of our cases can be considered as a biphenotypic leukemia according to the ranged from 1 to 20. The cytoplasm was scarce in five cases latter criteria. Nevertheless, in some cases to discard an imma- and abundant in three. In five cases, abundant mitochondria ture T cell acute lymphoblastic leukemia with some aberrant were seen and prominent Golgi apparatus and RER were seen myeloid antigens can be difficult. In this regard, none of our in one case. No ferritin particles, ␪ granules or rhopheocytosis patients showed cytoplasmic CD3 or a rearranged pattern for vesicles were observed. The nuclear outline was regular in C␤ or J␥ chain of TCR. 43% of the cases and irregular in 57%. Ultrastructural reaction As shown in our study, ultrastructural morphological study for platelet peroxidase was negative in the four cases tested. can be useful in the diagnosis of AML-M0 by revealing the Ultrastructural myeloperoxidase was positive, restricted to the presence of granules in the cytoplasm of the blast cells. Ultra- granules, in 12% of the blast cells in one out of the five structural MPO reaction was only positive in one of the five cases studied. cases tested with a granular pattern similar to that observed in other undifferentiated acute leukemias.23,24 The disagreement between immunological and ultrastructural MPO detection Discussion could be explained by technical differences (functional vs immunological detection) and the low percentage of positive The AML-M0 is a recently described AML subtype with a cells. It has been suggested that some AML-M0 could be of reported frequency ranging from 4% according to the FAB erythroid origin.15 There are no specific markers for very group1 to 10% found in a recent report.9 Only after the intro- immature erythroid cells, the ultrastructural analysis of the duction of immunophenotyping on a routine basis in the diag- cells being necessary for their detection.25 None of our nosis of acute leukemias has it been possible to identify the patients showed the presence of invaginated pinocytic ves- AML-M0 subtype. icles related to the intake of ferritin molecules at ultrastructural The reactivity with myeloid-specific MoAb is the main cri- level, thus excluding the presence of blasts of erythroid lin-

Table 4 Cytogenetic, molecular biology and ultrastructural characteristics of AML-M0

Patient 123456789

Ultrastructure Nucleus 76%I 60%I 56%R 80%R 56%R 36%I 48%I 80%I * Nucleoli 88% 80% 48% 60% 60% 88% 68% 87% * % granulated 80 56 20 36 40 36 4 40 * No. granules 2–20 1–14 1–2 1–9 1–10 1–9 1–2 1–3 * MPO 12% * * * −−−−* PPO − ***−−* − * Cytogenetics * 46XY * 46XY 46XX Hypotetr. 46XY Complex * Molecular biology

JH **RR**G** C␤ **GG**G** J␥ **GG**G** 3Јbcr **GG**G** Phenotypic and ultrastructural characteristics of AML-MO N Villamor et al 1075 eage. In addition, none of our cases expressed ultrastructural (AML-M0): comparison of 25 cases with other French–American– platelet peroxidase and were negative for the specific mega- British subtypes. Blood 1997; 89: 621–629. karyocytic lineage markers CD61 and CD41.25,26 12 Lepelley P, Preudhomme C, Sartiaux C, Ghevaert C, Lai JL, Iaru 10,16,27 T, Fenaux P, Cosson A. Immunological detection of myeloperoxid- Recent reports suggest that genetic abnormalities in ase in poorly differentiated acute leukemia. Eur J Hematol 1993; AML-M0 are frequently complex although without specific 50: 155–159. non-random abnormalities. Only one of our patients showed 13 Urbano-Ispizua A, Matutes E, Villamor N, Sierra J, Pujades A, a complex karyotype and a translocation with a breakpoint in Reverter JC, Feliu E, Cervantes F, Vives-Corrons JL, Montserrat E, 11(q23) that has been associated with MLL gene in acute Rozman C. The value of detecting surface and cytoplasmic anti- leukemias.28 gens in acute myeloid leukaemia. Br J Haematol 1992; 81: In conclusion, AML-M0 is an infrequent type of AML with 178–183. 14 Buccheri V, Shetty V, Yoshida N, Morilla R, Matutes E, Catovsky a phenotypical profile that can induce a misdiagnosis. The use D. The role of an anti-myeloperoxidase antibody in the diagnosis of monoclonal antibodies highly specific for each lineage, the and classification of acute leukaemia: a comparison with light and study of cytoplasmic antigens and the ultrastructural mor- electron microscopy cytochemistry. Br J Haematol 1992; 80: phology allow the confirmation of the myeloid origin and to 62–68. discard the presence of an immature erythroid or megakary- 15 Cuneo A, Ferrant A, Michaux JL, Boogaerts M, Demuynck H, van ocytic proliferation among AML-M0 cases. Orshoven A, Criel A, Stul M, Dal Clin P, Hernandez J, Chatelain B, Doyen C, Louwagie A, Castoldi G, Cassiman JJ, Van Den Ber- ghe H. Cytogenetic profile of minimally differentiated (FAB M0) acute myeloid leukemia: correlation with clinicobiological find- Acknowledgements ings. Blood 1995; 85: 3688–3694. 16 Stasi R, del Poeta G, Venditti A, Masi M, Stipa E, Dentamaro T, Cox C, Dallapiccola B, Papa G. Analysis of treatment failure in This work was partially supported by grants from Fondo de patients with minimally differentiated acute myeloid leukemia Investigaciones Sanitarias de la Seguridad Social, Ministerio (AML-M0). Blood 1994; 83: 1619–1625. de Sanidad (FIS 94/0839, 95/0257). 17 Drexler HG, Sperling C, Ludwig WD. Terminal deoxynucleotidyl tranferase (TdT) expression in acute myeloid leukemia. Leukemia 1993; 7: 1142–1150. 18 Drexler HG, Thiel E, Ludwig WD. Acute myeloid leukemias References expressing lymphoid-associated antigens: diagnostic incidence and prognostic significance. Leukemia 1993; 7: 489–498. 1 Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DAG, 19 Basan R, Biondi A, Benvestito S, Tini ML, Abbate M, Viero P, Gralnick HR, Sultan C. Proposal for the recognition of minimally Borleri G, Barbui T. Acute undifferentiated leukemia with CD7ϩ differentiated acute myeloid leukemia (AML-M0). Br J Haematol and CD13+ immunophenotype. Lack of molecular commitment 1991; 78: 325–329. and association with poor prognostic features. Cancer 1992; 69: 2 Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DAG, 396–404. ´ ´ ´ Gralnick HR, Sultan C. Criteria for the diagnosis of acute leukemia 20 Vidrales MB, Orfao A, Gonzalez M, Hernandez JM, Lopez-Berges of megakaryocyte lineage (M7). A report from the French–Amer- MC, Garcia MA et al. Expression of NK and lymphoid-associated ican–British cooperative group. Ann Intern Med 1985; 103: antigens in blast cells of acute myeloblastic leukemia. Leukemia 460–462. 1993; 12: 2026–2029. 3 Bennet JM, Catovsky D, Daniel MT, Flandrin G, Galton DAG, 21 Yokose N, Ogata K, Ito T, Miyake K, An E, Inokuchi K, Yamada Granick HR, Sultan C. Proposals for the classification of the acute T, Gomi S, Tanabe Y, Ohki I, Kuwabara T, Hasegawa S, Shinohara leukaemias. Br J Haematol 1976; 33: 451–458. T, Dan K, Nomura T. Chemotherapy for minimally differentiated 4 Bennet JM, Catovsky D, Daniel MT, Flandrin G, Galton DAG, acute myeloid leukemia (AML-M0). A report of five cases and Gralnick HR, Sultan C. Proposed revised criteria for the classi- review of the literature. Ann Hematol 1993; 66: 67–70. fication of acute myeloid leukemia. A report from the French– 22 Matutes E, Catovsky D. The value of scoring systems for the diag- American–British cooperative group. Ann Intern Med 1985; 103: nosis of biphenotypic leukemia and mature B-cell disorders. Leuk 620–625. Lymphoma 1994; 13 (Suppl. 1): 11–14. 5 Davey FR, Erber WN, Gatter KC, Mason DY. Immunophenotyping 23 Vainchenker W, Villeval JL, Tabilio A, Matamis H, Karianakis G, of acute myeloid leukemia by immunoalkaline phosphatase Guichard J et al. Immunophenotype of leukemic blasts with small (APAAP) labelling with a panel of antibodies. Am J Hematol 1987; peroxidase-positive granules detected by electron microscopy. 26: 157–166. Leukemia 1988; 2: 274–281. 6 Rozman C, Woessner S, Feliu E, Lafuente R, Berga L. Cell Ultras- 24 Heil G, Gunsilius E, Hoelzer D, Thiel E, Heimpel H. Peroxidase tructure for Hematologists. Doyma: Barcelona, 1993. expression in acute unclassified leukemias: ultrastructural studies 7 An International System for Human Cytogenetic Nomenclature. in combination with immunophenotyping. Leuk Lymphoma 1994; Recommendations of the International Standing Committee of 14: 103–109. Human Cytogenetic Nomenclature. Felix Mitelman (ed). Karger: 25 Breton-Gorius J. Review of the ultrastructural and cytochemistry of Basel, 1995. megakaryoblastic leukemia. In Poliack A (ed). Human Leukemias. 8 Maniatis T, Fritsch EF, Sambrook J. Molecular Cloning: A Labora- Martinus Nijhoff: Boston, 1984, pp 63–92. tory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1982. 26 Koike T. Megakaryoblastic leukemia: the characterization and 9 Venditti A, Del Poeta G, Stasi R, Masi M, Bruno A, Buccisano F, identification of megakaryoblasts. Blood 1984; 64: 683–692. Cox C, Coppetelli U, Aronica G, Simone MD, Tribaldo M, Ama- 27 Segeren CM, de Jong-Gerrits GCMM, van’t Veer MB. AML-MO: dori S, Papa G. Minimally differentiated acute myeloid leukemia clinical entity or waste basket for immature blastic leukemias? A (AML-M0): cytochemical, immunophenotypic and cytogenetic description of 14 patients. Ann Hematol 1995; 70: 297–300. analysis of 19 cases. Br J Haematol 1994; 88: 784–793. 28 Rubnitz JE, Behm FG, Downing JR. 11q23 rearrangements in acute leukemia. Leukemia 1996; 10: 74–82. 10 Venditti A, del Poeta G, Stasi R, Buccisano F, Aronica G, Bruno ˜ A, Cox C, Maffei L, Tamburini A, Papa G, Amadori S. Biological 29 Sierra J, Brunet S, Granena A, Olive T, Bueno J, Ribera JM, Petit profile of 23 cases of minimally differentiated leukemia (AML-M0) J, Besses C, Llorente A, Guardia R, Macia J, Rovira M, Badell I, and its clinical implication (letter). Blood 1996; 87: 418–420. Vela E, Diaz Heredia C, Vivancos P, Carreras E, Feliu E, Montser- 11 Venditti A, Del Poeta G, Buccisano F, Tamburini A, Cox C, Stasi rat E, Julia A, Cubells J, Rozman C, Domingo A, Ortega JJ. Feasi- R, Bruno A, Aronica G, Maffei L, Suppo G, Simone MD, Forte L, bility and results of bone marrow transplantation after remission Cordero V, Postorino M, Tufilli V, Isacchi G, Masi M, Papa G, induction and intensification chemotherapy in de novo acute Amadori S. Minimally differentiated acute myeloid leukemia myeloblastic leukemia. J Clin Oncol 1996; 14: 1353–1363.