Human Acute Myeloid Leukemia Cases Research Articles

In vivo genome-wide CRISPR screening in murine acute myeloid leukemia uncovers microenvironmental dependencies

AbstractGenome-wide CRISPR screens have been extremely useful in identifying therapeutic targets in diverse cancers by defining genes that are essential for malignant growth. However, most CRISPR screens were performed in vitro and thus cannot identify genes that are essential for interactions with the microenvironment in vivo. Here, we report genome-wide CRISPR screens in 2 in vivo murine models of acute myeloid leukemia (AML) driven by the KMT2A/MLLT3 fusion or by the constitutive coexpression of Hoxa9 and Meis1. Secondary validation using a focused library identified 72 genes specifically essential for leukemic growth in vivo, including components of the major histocompatibility complex class I complex, Cd47, complement receptor Cr1l, and the β-4-galactosylation pathway. Importantly, several of these in vivo–specific hits have a prognostic effect or are inferred to be master regulators of protein activity in human AML cases. For instance, we identified Fermt3, a master regulator of integrin signaling, as having in vivo–specific dependency with high prognostic relevance. Overall, we show an experimental and computational pipeline for genome-wide functional screens in vivo in AML and provide a genome-wide resource of essential drivers of leukemic growth in vivo.

Identification of two novel mutations in human acute myeloid leukemia cases

Acute myeloid leukemia (AML) is an aggressive cancer that progresses rapidly with a poor prognosis. Cytogenetic analysis provides the most accurate determination of diagnosis and prognosis however, about 42–48% of AML patients have a cytogenetically normal karyotype. Genetic analysis can provide further information and the identification of new mutations could result in improved risk stratification, prognosis and better understanding of the mechanisms of AML leukaemogenesis. In this study, we analyzed genetic alterations in 16 human AML cases by Haloplex sequencing with confirmation of two previously unreported mutations in the genes DNMT3A and RUNX1 by Sanger sequencing or pyrosequencing. The two novel mutations consist of two frameshift mutations identified in two different AML patients and reported as deleterious by bioinformatic analysis. These mutations confirm the exclusion and co-occurrence of specific gene mutation patterns in AML and may provide further information for patient diagnosis and prognosis.

The PARP Inhibitor Olaparib Antagonizes Leukemic Growth Induced By TET1 Overexpression in AML1-ETO Positive Acute Myeloid Leukemia

DNA methylation patterns are highly deregulated in human acute myeloid leukemia (AML) cases and stratify AML patient samples into different subgroup. AML1-ETO is the most commonly occurring fusion gene in AML and these AML cases exhibit an aberrant and distinct methylation pattern. So far, the underlying mechanisms for this are only poorly understood. The TET1 dioxygenase has recently emerged as an important epigenetic modifier: by catalyzing the conversion of 5-methylcytosine (5mC) to 5-hydroxymethylcytosine (5hmC) TET1 plays an important role in active demethylation, thereby regulating a variety of biological processes. It was linked to tumorigenesis based on the observation that its expression is frequently deregulated in solid cancer. However, the role of TET1 in AML1-ETO+ (AE+)human AML cases is yet unexplored. Using quantitative real time (qRT)- PCR we now show that AE+ AML is characterized by high and aberrant expression of TET1: the gene was significantly higher expressed in the majority of AE+ patients (n=7, p<0.01) compared to other AML subtypes such as inv(16) (n=11), PML-RARα+ (n=31), cytogenetically normal (CN)-AML patients (n=33) and CD34+ normal BM cells (n=4). This observation was consistent with published cDNA microarray data on large patient cohorts (Haferlach et al., JCO 2010, p<0.008 t-test, p<0.01 Anova) and recently published transcriptome data (TCGA) of AML patients. In contrast to TET1, TET2 and TET3 did not show significant higher expression in AE+ patients compared to other AML subtypes. In line with patient data, TET1 was highest expressed in the AE+ AML cell line KASUMI-1 and SKNO-1 compared to other AML cell lines (p<0.05 and n=3). Compared to normal CD34+ and myeloid (CD33+, CD15+ and CD14+) cells (n=3), TET1 was 10-fold and 16-fold higher expressed in AE+ patient samples (n= 7). Aberrant expression of TET1 in AE+ leukemic cells was associated with hypomethylation of its promoter and enrichment for H3K4me3 euchromatic marks at its promoter as determined by LC/MS and ChIP-qPCR respectively. Knockdown (KD) of TET1 mRNA using two short hairpin RNAs (shRNAs) in AE+ AML cell lines impaired their cell growth and clonogenicity by over 50% in vitro (n=3 and p<0.01). shRNA mediated depletion of TET1 did not impact the cell growth and clonogenicity of the TET1 negative cell line RAJI, ruling out off target effects of the shRNAs (n=3). In mice, KD of Tet1 in leukemic bone marrow cells expressing the truncated leukemogenic AML1-ETO9a (AE9a) fusion, dramatically inhibited cell growth (>60% compared to scrambled, n=3, p<0.01), clonogenicity (>50-70% reduction in primary CFCs, p<0.01, n=3) and importantly delayed onset of leukemia in vivo (median survival 35 days for scr vs 80 days for shRNA mice, n=4/arm, p<0.03). Tet1-knock-out c-kit+ hematopoietic stem and progenitor cells (HSPCs) transduced with AE9a showed reduced primary colony formation and impaired serial replanting capacity in vitro compared to AE9a transduced Tet1-wild-type HSPCs (>50% and >70%, respectively; p<0.001, n=3). Global analysis of 5hmC and 5mC levels using hMeDIP/MeDIP-seq performed on TET1 depleted KASUMI-1 cells revealed lower global 5hmC levels and increase in 5mC as compared to cells transduced with scrambled control (n=2). 3324 promoter regions lost 5hmC and gained 5mC upon TET1 depletion (-5kTSS, Fold enrichment cut off <2-fold, q-value<1e5). Recent studies have shown that PARP activity induces TET1 expression by regulating its promoter epigenetically. We could show that aberrant TET1 expression could be antagonized by the PARP inhibitor olaparib in AE+ leukemic cell lines. Furthermore, olaparib treatment decreased 5hmC levels and reduced cell growth and clonogenicity of human AE+ cell lines and of the murine AE9a+ leukemic cell line in vitro (n=3, p<0.01). In conclusion, our data indicates that aberrant TET1 expression contributes to the growth of AE+ AML by maintaining the 5-hydroxymethylome and that the PARP inhibitor olaparib can at least partially antagonize the oncogenic effect of TET in AML. DisclosuresMulaw:NuGEN: Honoraria. Buske:Celltrion, Inc.: Consultancy, Honoraria.

The +37 kb Cebpa Enhancer Is Critical for Cebpa Myeloid Gene Expression and Contains Functional Sites that Bind SCL, GATA2, C/EBPα, PU.1, and Additional Ets Factors.

The murine Cebpa gene contains an evolutionarily conserved 453 bp enhancer located at +37 kb that, together with its promoter, directs expression to myeloid progenitors and to long-term hematopoietic stem cells in transgenic mice. In human acute myeloid leukemia cases, the enhancer lacks point mutations but binds the RUNX1-ETO oncoprotein. The enhancer contains the H3K4me1 and H3K27Ac histone modifications, denoting an active enhancer, at progressively increasing levels as long-term hematopoietic stem cells transition to granulocyte-monocyte progenitors. We previously identified four enhancer sites that bind RUNX1 and demonstrated that their integrity is required for maximal enhancer activity in 32Dcl3 myeloid cells. The +37 kb Cebpa enhancer also contains C/EBP, Ets factor, Myb, GATA, and E-box consensus sites conserved in the human +42 kb CEBPA enhancer. Mutation of the two C/EBP, seven Ets, one Myb, two GATA, or two E-box sites reduces activity of an enhancer-promoter reporter in 32Dcl3 cells. In 293T gel shift assays, exogenous C/EBPα binds both C/EBP sites, c-Myb binds the Myb site, PU.1 binds the second Ets site, PU.1, Fli-1, ERG, and Ets1 bind the sixth Ets site, GATA2 binds both GATA sites, and SCL binds the second E-box. Endogenous hematopoietic RUNX1, PU.1, Fli-1, ERG, C/EBPα, GATA2, and SCL were previously shown to bind the enhancer, and we find that endogenous PU.1 binds the second Ets site in 32Dcl3 cells. Using CRISPR/Cas9, we developed 32Dcl3 lines in which the wild-type enhancer alleles are replaced with a variant mutant in the seven Ets sites. These lines have 20-fold reduced Cebpa mRNA when cultured in IL-3 or G-CSF, demonstrating a critical requirement for enhancer integrity for optimal Cebpa expression. In addition, these results indicate that the +37 kb Cebpa enhancer is the focus of multiple regulatory transcriptional pathways that impact its expression during normal hematopoiesis and potentially during myeloid transformation.

Knockdown of SALL4 Protein Enhances All-trans Retinoic Acid-induced Cellular Differentiation in Acute Myeloid Leukemia Cells

All-trans retinoic acid (ATRA) is a differentiation agent that revolutionized the treatment of acute promyelocytic leukemia. However, it has not been useful for other types of acute myeloid leukemia (AML). Here we explored the effect of SALL4, a stem cell factor, on ATRA-induced AML differentiation in both ATRA-sensitive and ATRA-resistant AML cells. Aberrant SALL4 expression has been found in nearly all human AML cases, whereas, in normal bone marrow and peripheral blood cells, its expression is only restricted to hematopoietic stem/progenitor cells. We reason that, in AMLs, SALL4 activation may prevent cell differentiation and/or protect self-renewal that is seen in normal hematopoietic stem/progenitor cells. Indeed, our studies show that ATRA-mediated myeloid differentiation can be largely blocked by exogenous expression of SALL4, whereas ATRA plus SALL4 knockdown causes significantly increased AML differentiation and cell death. Mechanistic studies indicate that SALL4 directly associates with retinoic acid receptor α and modulates ATRA target gene expression. SALL4 is shown to recruit lysine-specific histone demethylase 1 (LSD1) to target genes and alter the histone methylation status. Furthermore, coinhibition of LSD1 and SALL4 plus ATRA treatment exhibited the strongest anti-AML effect. These findings suggest that SALL4 plays an unfavorable role in ATRA-based regimes, highlighting an important aspect of leukemia therapy.

Abstract 1108: Global discovery of dysregulated protein expression and phosphorylation networks identifies Leo1 as a key substrate of PRL-3 phosphatase in leukemogenesis.

Abstract Cytogenetically normal acute myeloid leukemia (AML) constitutes about 40-50% of human AML cases, is a highly heterogeneous and poorly understood. Identification of novel oncogenic alterations and pathways are the key to the better understanding and discovery of useful therapeutic targets in the improved management of AML. PRL-3 is a small 20kDa prenylated dual-specificity phosphatase that is over-expressed in about 50% of acute myeloid leukemia (AML) and associated with poorer survival, but the mechanisms regulating the associated malignancy remains poorly understood. Our in vitro data demonstrates that ectopic PRL-3 expression in the factor-dependent TF1 AML cells confers cytokine-independent growth, induces colony-forming ability in methylcellulose media and tumorigenesis in vivo. To systematically characterize novel substrates of PRL-3 in leukemia, unbiased large-scale proteomics and phosphoproteomics analyses were performed between the parental TF1 cells and their malignant PRL-3 transfectant counterparts to discover critical differences in signaling networks. We obtained quantitative measurements on 803 proteins, where 331 were significantly up-regulated (&amp;gt;1.5-fold) and 67 were under-expressed (&amp;lt;0.6-fold). More importantly, PRL-3 alters the phosphorylation status of 192 proteins. Our proteomics profiling and in vitro validation data revealed that Leo1, component of the Polymerase II-associating factor 1 (PAF) complex, is upregulated and dephosphorylated by PRL-3. Mechanistically, PRL-3 regulates Leo1 by relieving the repressive H3K9me3 mark on Leo1 promoter through direct binding of JMJD2C histone demethylase, promoting transcriptionally permissive chromatin state leading to efficient Leo1 gene expression. Conversely, abrogation of PRL-3 using RNAi reduced Leo1 levels and destabilizes the PAF complex, consequently leading to the downregulation of PAF-regulated pluripotency target genes like Sox2, Sox4 and Tbx3. Our functional analyses corroborated the contribution of Leo1 towards PRL-3 oncogenicity, whereby inhibition of Leo1 in PRL-3 cells impeded cell proliferation, induced significant apoptosis and abolished colony formation. Finally, we relate these data to clinical relevance where 46% of the human AML patient samples and 45.5% of the AML cell lines showed a positive correlation between PRL-3 expression and abnormal Leo1 activation. In conclusion, our work allows the identification of bona fide PRL-3 targets and signaling networks, which established starting points for a complete enumeration of PRL-3 function in AML signaling. The improved understanding of this network provides valuable leads for designing future therapies targeting PRL-3 in leukemia and carcinomas. Citation Format: Phyllis SY Chong, Jian Biao Zhou, Tiannan Guo, Siu Kwan Sze, Qi Zeng, Wee Joo Chng. Global discovery of dysregulated protein expression and phosphorylation networks identifies Leo1 as a key substrate of PRL-3 phosphatase in leukemogenesis. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 1108. doi:10.1158/1538-7445.AM2013-1108

Evi-1 is a transcriptional target of mixed-lineage leukemia oncoproteins in hematopoietic stem cells

AbstractEcotropic viral integration site-1 (Evi-1) is a nuclear transcription factor that plays an essential role in the regulation of hematopoietic stem cells. Aberrant expression of Evi-1 has been reported in up to 10% of patients with acute myeloid leukemia and is a diagnostic marker that predicts a poor outcome. Although chromosomal rearrangement involving the Evi-1 gene is one of the major causes of Evi-1 activation, overexpression of Evi-1 is detected in a subgroup of acute myeloid leukemia patients without any chromosomal abnormalities, which indicates the presence of other mechanisms for Evi-1 activation. In this study, we found that Evi-1 is frequently up-regulated in bone marrow cells transformed by the mixed-lineage leukemia (MLL) chimeric genes MLL-ENL or MLL-AF9. Analysis of the Evi-1 gene promoter region revealed that MLL-ENL activates transcription of Evi-1. MLL-ENL–mediated up-regulation of Evi-1 occurs exclusively in the undifferentiated hematopoietic population, in which Evi-1 particularly contributes to the propagation of MLL-ENL–immortalized cells. Furthermore, gene-expression analysis of human acute myeloid leukemia cases demonstrated the stem cell–like gene-expression signature of MLL-rearranged leukemia with high levels of Evi-1. Our findings indicate that Evi-1 is one of the targets of MLL oncoproteins and is selectively activated in hematopoietic stem cell–derived MLL leukemic cells.

Differential Contributions of c-Kit Activating Mutations to Promotion of AML1-ETO Associated Neoplasia

AbstractAbstract 4195The t(8;21) AML1-ETO translocation is one of the most frequent cytogenetic abnormalities observed in acute myeloid leukemia (AML), occurring in 10–12% of cases. The t(8;21) fuses the DNA binding domain of AML1 with the transcriptional repressor gene ETO. AML1-ETO dominantly blocks wild-type AML1 function, preventing activation of AML1 target genes and impairing myeloid differentiation. However, expression of AML1-ETO in mice does not promote AML in the absence of additional mutations. Another commonly mutated gene detected in AML1-ETO+ patient samples is the receptor tyrosine kinase c-KIT. Two classes of activating c-KIT mutations have been reported; one where mutations like D814V occur in the tyrosine kinase domain and the other involving a deletion within Exon 8 (E8D419) in a region that mediates receptor dimerization. These mutations are not randomly distributed with t(8;21) or with other chromosomal translocations found in AML. For instance, c-KITD814V is observed in 11–44% of t(8;21) patients, whereas c-KITE8Æ419 with t(8;21) is reported in 2–13% of cases. These observations suggest a degree of specificity with which activating c-KIT mutations can cooperate with AML1-ETO to promote AML. To test this hypothesis, we used a retroviral transduction/transplantation model to co-express AML1-ETO with c-KitD814V or c-KitE8Æ419 in murine bone marrow (BM) progenitor cells used to reconstitute lethally irradiated mice. Analysis of reconstituted mice has shown that simultaneous expression of AML1-ETO with c-KitD814V results in three non-overlapping phenotypes. The most common phenotype, occurring in 50% of animals, is a transplantable AML characterized by a relatively short disease latency that may or may not present with an associated granulocytic sarcoma. Second, a rapidly fatal non-transplantable myeloproliferative phenotype was observed in 25% of animals. Lastly, several mice (25%) developed a lethal, short latency B-cell leukemia that was transplantable to secondary recipient animals. The heterogeneity of disease phenotypes that develop in AML1-ETO;c-KITD814V mice may reflect the specific hematopoietic progenitor cell type that was initially transduced and capable of reconstituting the recipient mouse or differences in AML1-ETO and/or c-KitD814V expression levels. In contrast, co-expression of AML1-ETO with c-KitE8Æ419 promoted a transplantable AML phenotype characterized by a relatively long disease latency in a fraction (43.7%) of reconstituted mice. AML1-ETO; c-KitE8Æ419–expressing cells could persist for months without showing signs of selective expansion, which suggests that additional genetic changes are necessary for overt transformation to AML in this context. These observations indicate that when overlaid on the AML1-ETO genetic lesion, c-KitD814V promotes a more aggressive disease phenotype than c-KitE8Æ419, which displays a longer latency and reduced penetrance. These differences may account for the observed disparity in the frequencies of each c-KIT mutation concurrent with AML1-ETO in human AML cases. Disclosures:No relevant conflicts of interest to declare.

Mutations of the TET2 and CBL genes: novel molecular markers in myeloid malignancies

Despite recent progress in molecular research in myeloid malignancies, in subsets of patients with myelodysplastic syndrome (MDS) so far no underlying mutation was identified. In the myeloproliferative neoplasms (MPNs), the JAK2V617F alone cannot explain the phenotypic heterogeneity. In acute myeloid leukemia (AML), clinical variability exists within distinct subgroups. Thus, the search for novel molecular markers continues. Recently, mutations of the tet oncogene family member 2 (TET2) and Casitas B-cell lymphoma (CBL) genes became the focus of interest. With diverse genetic methods, TET2 on chromosome 4q24 was identified as candidate tumor suppressor gene. Sequencing studies revealed heterogeneous mutations in 10-25% of patients with acute myeloid leukemia (AML), MDS, and MPNs, while the frequency might be higher in chronic myelomonocytic leukemia (CMML). The prognostic impact is being explored. The CBL gene is involved in the degradation of tyrosine kinases. In rare cases of human AML (<2%), CBL mutants were identified, with a higher frequency in core binding factor leukemias. Presence of these mutations was suggested to be involved in aberrant FLT3 expression. In the MPNs, a 2-8% frequency of CBL mutations was reported. These novel mutations deepened insights in the mechanisms of leukemogenesis, might contribute to the identification of new therapeutic targets, and improve diagnostics in the myeloid malignancies.

Identification, characterization, and function of a novel oncogene: the peripheral cannabinoid receptor Cb2.

By means of retroviral insertional mutagenesis, we identified a novel common virus integration site (cVIS) (Evi11) in murine leukemias, and demonstrated that Cb2, encoding the peripheral cannabinoid receptor, is the potential proto-oncogene located in that region. Cb2 is a 7-transmembrane G protein-coupled receptor (GPCR), which is normally expressed on B lymphocytes. Using transwell assays we observed strong migration of Cb2-expressing cells upon stimulation with 2-arachidonoylglycerol (2-AG), a potent endocannabinoid. Overexpression of Cb2 receptor on murine myeloid precursor cells causes a block of neutrophilic differentiation, a major characteristic of myeloid leukemia. Intriguingly, we could not detect functional Cb2 receptors on normal murine bone marrow precursor cells. Furthermore, analysis of human acute myeloid leukemia (AML) samples revealed the presence of CB2 mRNA transcripts in several cases. Furthermore, migration could be induced by 2-AG when analyzed in one of the patient samples. Our data suggest that the initially identified cVIS, Evi11, encodes for a murine onco-protein and that human CB2 may be involved in certain cases of human AML as well.

Cytokine upregulation of the antigen presenting function of acute myeloid leukemia cells.

Acute myeloid leukemia (AML) cells are malignant counterparts of normal myeloid pathway progenitors. Myeloid progenitors differentiate into professional antigen presenting cells (APC) under the essential influence of GM-CSF along with additional cytokines. Twelve cases of human AML were tested for ability to be differentiated toward a professional APC phenotype in short-term culture with addition of GM-CSF and the following recombinant proteins: TNFalpha, IL-4, CD40 ligand, Flt3 ligand and SCF. Significant upregulation of CD80 (B7-1) and enhancement of alloantigen presentation was seen with the addition of GM-CSF and TNFalpha alone or with additional cytokines. The combination of GM-CSF and TNFalpha, either alone or in combination with an additional cytokine, resulted in enhancing alloantigen presentation by at least two-fold over the media control group in 10/12 patients studied, and resulted in CD80 expression of greater than 15% in 11/12 patients studied. In AML cultures with GM-CSF and TNFalpha, coexpression of CD80 and either CD34 or an aberrant surface marker (CD56) was seen. In one case, sorted CD80, cells retained a characteristic cytogenetic marker and CD34 expression, proving their derivation from an AML precursor. These studies verify other reports of in vitro differentiation of human AML precursors into enhanced APC, suggesting that this phenomenon could be utilized for immunotherapy strategies aimed at enhancing presentation of leukemia antigens to T cells.

Growth regulation of human acute myeloid leukemia: effects of five recombinant hematopoietic factors in a serum-free culture system

The response of human acute myeloid leukemia (AML) cells to the distinct hematopoietic growth factors (HGFs), ie, recombinant interleukin-3 (IL-3), granulocyte-macrophage-CSF (GM-CSF), granulocyte- CSF (G-CSF), macrophage-CSF (M-CSF), and erythropoietin (Epo) was investigated under well-defined serum-free conditions. Proliferative responses to these factors, when added separately as well as in combinations, were analyzed in 25 cases of human AML using 3H-thymidine incorporation and colony assays. The 3H-thymidine uptake data revealed that IL-3, GM-CSF, G-CSF, and M-CSF were stimulators of AML proliferation in 19, 15, 13, and 4 cases, respectively. Epo only stimulated DNA synthesis in the cells of the single erythroleukemia case. GM-CSF stimulation was seen only in IL-3 reactive cases and GM- CSF, when combined with IL-3, could not further elevate the DNA synthesis evoked by IL-3 alone. On the other hand, in six cases, G-CSF enhanced the IL-3- or GM-CSF-stimulated thymidine uptake. These results suggest that subpopulations of AML cells that are activated by distinct CSFs (eg, IL-3/GM-CSF-responsive cells and G-CSF-responsive cells) coexist. The 3H-thymidine incorporation assay was more sensitive for measuring CSF responses than methylcellulose colony cultures, since activation of DNA synthesis was more frequently seen than induction of colony formation. DNA synthesis experiments revealed eight different CSF response patterns among these 25 cases. CSF phenotyping may be a useful addition to the morphologic classification of AML, since these patterns directly reflect the ability of the proliferating subsets of AML cells to respond to the CSFs.

Growth Regulation of Human Acute Myeloid Leukemia: Effects of Five Recombinant Hematopoietic Factors in a Serum-Free Culture System

The response of human acute myeloid leukemia (AML) cells to the distinct hematopoietic growth factors (HGFs), ie, recombinant interleukin-3 (IL-3), granulocyte-macrophage-CSF (GM-CSF), granulocyte-CSF (G-CSF), macrophage-CSF (M-CSF), and erythropoietin (Epo) was investigated under well-defined serum-free conditions. Proliferative responses to these factors, when added separately as well as in combinations, were analyzed in 25 cases of human AML using 3H-thymidine incorporation and colony assays. The 3H-thymidine uptake data revealed that IL-3, GM-CSF, G-CSF, and M-CSF were stimulators of AML proliferation in 19, 15, 13, and 4 cases, respectively. Epo only stimulated DNA synthesis in the cells of the single erythroleukemia case. GM-CSF stimulation was seen only in IL-3 reactive cases and GM-CSF, when combined with IL-3, could not further elevate the DNA synthesis evoked by IL-3 alone. On the other hand, in six cases, G-CSF enhanced the IL-3- or GM-CSF-stimulated thymidine uptake. These results suggest that subpopulations of AML cells that are activated by distinct CSFs (eg, IL-3/GM-CSF-responsive cells and G-CSF-responsive cells) coexist. The 3H-thymidine incorporation assay was more sensitive for measuring CSF responses than methylcellulose colony cultures, since activation of DNA synthesis was more frequently seen than induction of colony formation. DNA synthesis experiments revealed eight different CSF response patterns among these 25 cases. CSF phenotyping may be a useful addition to the morphologic classification of AML, since these patterns directly reflect the ability of the proliferating subsets of AML cells to respond to the CSFs.