Molecular Biology of Myelodysplasia Philip Nivatpumin, Steven Gore

Contents

4.1 Introduction . . . . 23

4.2 Pathogenesis . . . . 24

4.2.1 Susceptibility to Genomic Injury/ Genomic Instability . . . . 24

4.2.2 Epigenetic Modifications . . . . 24

4.2.3 Abnormalities in Signaling Pathways 25 4.2.3.1 EPO/Growth Factor Signaling 25 4.2.3.2 RAS/MAPK . . . . 25

4.2.3.3 Vascular Endothelial Growth Factor (VEGF) . . . . 26

4.2.4 Apoptosis . . . . 26

4.2.5 Immune Dysregulation . . . . 27

4.2.6 Bone Marrow Microenvironment . . 28

4.2.6.1 Cytokine Milieu . . . . 28

4.2.6.2 Neo-angiogenesis . . . . 28

4.2.7Molecular Abnormalities Identified by Cytogenetic Features . . . . 29

4.2.7.1 Chromosome 5 Deletions . 29 4.2.7.2 Chromosome 7 Deletions . 29 4.2.7.3 Chromosome 20 . . . . 30

4.2.7.4 Chromosome 17 . . . . 30

4.2.7.5 Trisomy 8 . . . . 30

4.2.7.6 Other Less Frequent Chro- mosomal Deletions . . . . 30

4.2.8 Translocations . . . . 30

4.2.8.1 TEL(ETV6) Fusion . . . . 30

4.2.8.2 Nucleoporin Abnormality . . 30

4.2.8.3 MLL . . . . 31

4.2.8.4 EVI-1 . . . . 31

4.2.8.5 NPM . . . . 31

4.2.9 Molecular Abnormalities Unrelated to Cytogenetic Abnormalities . . . 31

4.2.9.1 TP53 . . . . 31

4.2.9.2 FLT3 . . . . 31

4.2.9.3 AML1 . . . . 32

4.2.9.4 Other Mutations . . . . 32

4.2.10 Gene Expression Profiling . . . . 32

4.3 Summary . . . . 32

References . . . . 33

4.1 Introduction

The myelodysplastic syndromes (MDS) are a heteroge- neous group of monoclonal disorders characterized by ineffective hematopoiesis and an increased risk of trans- formation to acute leukemia. Current classification sys- tems (e.g., French-American-British (FAB), World Health Organization (WHO)) are based on morpholog- ical features of blood and bone marrow elements as well as cytogenetic abnormalities (Bennett et al. 1982; Harris et al. 1999). The natural history of MDS is highly vari- able. This likely reflects the myriad cytogenetic, genetic and epigenetic alterations that are associated with MDS.

It has generally been suggested that MDS arises from a

hematopoietic stem cell that has suffered irreversible

DNA damage. Further events result in dominance of this

damaged clone. Immunologic responses may occur that

promote progenitor survival and eventual clonal domi-

nance. In bone marrows from early-stage MDS patients,

impaired differentiation and increased apoptosis pre-

Philip Nivatpumin, Steven Gore

dominate, but as disease progression occurs, increased proliferation and accumulation of immature cells result.

This chapter will summarize some of the observations of the numerous biologic mechanisms implicated in the pathophysiology of MDS, including intrinsic pro- genitor abnormalities, epigenetic changes, abnormal apoptosis machinery, immunologic influences, abnor- mal signal transduction pathways and the role of the bone marrow microenvironment.

4.2 Pathogenesis

4.2.1 Susceptibility to Genomic Injury/Genomic Instability

MDS is thought to arise from the somatic mutation of a hematopoietic progenitor cell. Confirmation of the clo- nality has been bolstered by analysis of cytogenetic and X-chromosome inactivation studies (Abrahamson et al.

1991; Delforge 2003) of patients with MDS. However, sci- entific evidence has grown in support of the concept that the cytogenetic abnormalities seen so frequently in MDS may be acquired during disease progression, rather than reflecting the initial inciting clonal event (Delforge 2003; Nilsson et al. 2002). Whether a primary or secondary event, genomic instability, as evidenced by karyotypic changes common in MDS, is thought to play an important role in disease pathogenesis.

Cytogenetic abnormalities in MDS result from the accumulation of genomic damage, failure to repair such damage, or both. Although the etiology of most cases of MDS is unknown, exposure to genotoxic agents such as benzene, radiation, or prior treatment with chemother- apeutic agents is known to increase the risk of develop- ing MDS (Nisse et al. 2001). Other environmental agents that may increase the risk include smoking, heavy met- als, pesticides, fertilizers, petroleum products, and or- ganic chemicals (Garfinkel and Boffetta 1990; Rigolin et al. 1998; West et al. 1995) (see Chapter 3).

The relationship between Fanconi anemia (FA) and childhood MDS point to increased susceptibility to genomic damage as an important cause of childhood MDS. Fanconi anemia is an autosomal recessive disor- der of chromosomal instability. Patients are character- ized by congenital abnormalities, ineffective hematopoi- esis, and a high risk of developing MDS, acute leukemia, and solid tumors. FA proteins interact with other DNA damage repair proteins such as ATM and BRCA1 and

BRCA2 (Tischkowitz and Dokal 2004). Repair of DNA double-strand breaks is often inaccurate in the pre-leu- kemic Bloom's syndrome and in FA (Gaymes et al. 2002;

Langland et al. 2002). Impaired response to oxidative stress has also been implicated in the pathophysiology of MDS. Defects in both glutathione transferase theta 1 (GSTT1) and NADPH quinone oxyreductase (NQO1) have been associated with an increased risk of myelo- dysplastic syndrome (Chen et al. 1996; Farquhar and Bowen 2003; Rothman et al. 1997). Glutathione is both an antioxidant and a cofactor for many antioxidant en- zymes that are important in the metabolism of various toxins and carcinogens. NQO1 is another gene involved in antioxidant mechanisms. Mutations in this gene have been reported in association with benzene-induced bone marrow damage.

Another possible mechanism underlying genomic instability involves telomere dynamics and the enzyme telomerase. Telomere erosion may result in chromo- some end fusion and subsequent chromosome instabil- ity. Shortened telomere length has been reported to be associated with poor prognosis in patients with MDS (Engelhardt et al. 2004; Ohyashiki et al. 1994).

4.2.2 Epigenetic Modifications

While genetic alterations are critical in the pathogenesis of MDS, epigenetic changes also contribute significantly to the disease phenotype. A modern definition of epige- netics refers to ªmodifications in gene expression that are brought about by heritable, but potentially revers- ible changes in chromatin structure and/or DNA meth- ylationº (Henikoff and Matzke 1997). Epigenetic changes include methylation of cytosine residues followed by a guanine base (DNA methylation) and post-translational modifications of histones that lead to alteration in chro- matin structure at specific gene loci, which in turn de- termine the transcriptional output of the gene. These changes to histones, collectively referred to as the his- tone code, include lysine acetylation, lysine methyla- tion, ubiquitination, phosphorylation, and sumoylation.

These histone modifications are recognized by specific proteins that recruit transcriptional activators and co- repressors, establishing a higher order of chromatin structure (Fischle et al. 2003; Hake et al. 2004).

Methylation of CpG dinucleotides concentrated in

the promoter regions of some genes (so-called `CpG is-

lands') results in the functional inactivation of those

genes without alteration in the primary sequence. Both hypomethylation and hypermethylation of the genome have been observed in hematological malignancies. Hy- pomethylation of the MDR1 gene has been described in AML samples and was shown to correlate with in- creased expression as measured by reverse transcrip- tion polymerase chain reaction (RT-PCR), possibly con- tributing to multidrug resistance in these patients (Att- wood et al. 2002; Nakayama et al. 1998). Hypomethyla- tion of c-myc and myeloperoxidase have also been de- scribed in samples from patients with primary AML and AML arising from MDS (Tsukamoto et al. 1992).

While mutations in cell cycle control genes such as p15, p16, and p19 have rarely been described in MDS, hy- permethylation of p15 is common. Hypermethylation of the p15

gene promoter has been observed in 30±

50% of MDS cases and has been shown to correlate with the percentage of bone marrow blasts (Quesnel et al.

1998; Uchida et al. 1997). p15

is a cyclin-dependent kinase inhibitor that is critical in regulating the G1 phase of the cell cycle. Its activation is downstream of the TGF-b/SMAD pathway. This suggests that one mech- bb anism of proliferation of leukemic cells is escape of reg- ulation of G1 phase of cell cycle. Further evidence for the importance of this event in MDS pathogenesis de- rives from the observation that the degree of methyla- tion correlates with the risk of evolution to AML and clinical prognosis (Quesnel et al. 1998). Other genes fre- quently methylated and silenced in myeloid malignan- cies include E-Cadherin, RARb R , and SOCS-1 (Esteller and Herman 2002; Herman and Baylin 2003).

The clinical activity of the DNA methyltransferase inhibitors 5-azacitidine (Silverman 2004) and 2'- deoxy-5-azacitidine (Wijermans et al. 1997, 2000) in MDS suggests that the methylation status of a subset of genes is likely to contribute significantly to the bio- logical and clinical behavior of MDS. However, attempts to correlate the clinical activity of these agents with re- versal of p15 methylation have demonstrated that meth- ylation reversal may not be required for clinical re- sponse (Daskalakis et al. 2002; Issa et al. 2004; Lubbert 2003).

Transcriptional silencing of methylated genes is mediated at least in part through the establishment of repressive chromatin conformation through the recruit- ment of histone deacetylases (Cameron et al. 1999). This has led to the strategy of combined DNA methyltransfer- ase and histone deacetylase inhibition for the treatment of MDS.

4.2.3 Abnormalities in Signaling Pathways 4.2.3.1 EPO/Growth Factor Signaling

The ineffective hematopoiesis that is characteristic of MDS has led to the investigation of pathways involved in transducing signals from erythropoietin (EPO) and other growth factors. When bone marrow cells from MDS patients are cultured in colony forming assays in the presence of EPO, erythroid colony formation is re- duced in comparison to normal controls (Backx et al.

1993; Mayani et al. 1989). EPO signaling involves a com- plex cascade of events beginning with the binding of EPO to the erythropoietin receptor (EPO-R). Upon binding EPO, the Janus kinase, JAK2, is activated (Tan- ner et al. 1995; Witthuhn et al. 1993). JAK2 activation leads to downstream tyrosine phosphorylation of a number of proteins. Activation of signal transducer and activator of transcription 5 (STAT5), a downstream protein of JAK2 that is thought to be important in EPO signaling, is impaired in myelodysplastic syndrome (Hoefsloot et al. 1997). This observation, combined with the other reports of normal presence of EPO-R in MDS patients, indicates that alteration of the EPO signaling pathway may have an important role in MDS (Backx et al. 1996).

Alterations in other growth factor pathways have also been reported in MDS patient samples. GM-CSF and G-CSF priming of reactive oxygen species (ROS) production in neutrophils of patients with MDS is im- paired (Fuhler et al. 2003). Thrombopoietin signaling has been investigated for its role in the dysmegakaryo- cytopoiesis seen in MDS, but its role is unclear (Hof- mann et al. 2000; Kalina et al. 2000).

4.2.3.2 RAS/MAPK

RAS is a critical component in the signaling cascade re- sulting in cellular proliferation in response to a variety of extracellular signals including growth factors. It acts as a molecular relay switch that is downstream of recep- tor tyrosine kinases and upstream of a cascade of mito- gen activated protein kinases (MAPK) that result in ac- tivation of nuclear transcription factors. RAS mutations have been identified in up to 30% of human leukemias and up to 15% of patients with MDS (Paquette et al.

1993). N-ras mutations are associated with poor prog-

nosis in MDS patients and increased rate of transforma-

tion to AML (Paquette et al. 1993). RAS activation is

mediated by the hydrolysis of GTP. RAS has intrinsic GTP-binding activity and GTP-hydrolyzing activity.

This RAS/GTP complex activates target proteins such as RAF, leading to subsequent downstream signaling.

Point mutations that interfere with the GTP hydrolyzing activity of RAS result in constitutive signaling and acti- vation of downstream components, resulting in cell pro- liferation. An N-ras mutation has been associated with an increased risk of progression to AML (Padua et al.

1998). Preliminary clinical studies suggest that the far- nesyl transferase inhibitors (FTI), tipifarnib and lona- farnib, are active in MDS; however, it is not clear that the clinical activity is related to the impact of the FTIs on RAS molecular signaling (Kantarjian et al. 2002;

Kurzrock et al. 2004).

4.2.3.3 Vascular Endothelial Growth Factor (VEGF)

Vascular endothelial growth factor (VEGF) is a key reg- ulator of angiogenesis. Angiogenesis is increasingly thought to have an important role in MDS (see Angio- genesis below). VEGF is regulated by multiple signals, including hypoxia inducible factor-1 (HIF-1) and Ras (Estey 2004). Myelomonocytic precursors of patients with MDS and AML overexpress both VEGF and its high affinity receptor, Flt-1 (Bellamy et al. 2001). Inhibi- tion of VEGF reduces leukemia colony formation in clinical samples from MDS patients (Broxmeyer et al.

1995). The importance of VEGF has led to the clinical investigation of VEGF receptor antibodies and VEGF tyrosine kinase inhibitors for the treatment of MDS.

4.2.4 Apoptosis

Apoptosis is an ordered cellular process that regulates cell population size in a variety of conditions. First de- scribed by Wyllie and colleagues (Wyllie et al. 1980), apoptosis is an energy-dependent process characterized morphologically by cytoplasmic and nuclear condensa- tion, fragmentation of nuclei into ªapoptotic bodiesº, preservation of plasma membrane integrity and phago- cytosis of cellular debris by macrophages in the absence of an inflammatory response (Greenberg 1998; Walker et al. 1988; Wyllie 1985). This death mechanism is crucial in maintaining a precise number of cells in a given or- ganism. Alterations in apoptosis have been implicated in a variety of medical disorders including myelodyspla-

sia. A variety of stimuli or insults serve as initiators of the apoptotic pathway. These include chemotherapy drugs, ultraviolet and gamma irradiation, chemical ex- posure, viral infection, steroid hormones, and various cytokines (e.g., TNF-a, Fas ligand, TGF-b) (Greenberg 1998).

A comprehensive review of all the pathways involved in apoptosis is outside the scope of this chapter. However, the process of apoptosis may be conceptually divided into extrinsic and intrinsic pathways. Extrinsic triggers include death ligands (e.g., Fas ligand, TNF-a, TNF-re- lated apoptosis-inducing ligand [TRAIL]) which bind to cell surface receptors and activate downstream signal transduction pathways. Intrinsic signals that activate the apoptotic pathway result from cellular stress, including exposure to radiation, chemicals or infectious processes.

Removal of cellular survival signals (e.g., growth factors) may also trigger intrinsic activation of cell death. Both extrinsic and intrinsic apoptotic pathways culminate in the activation of a family of cytosolic aspartate-specific cysteine proteases (caspases). Caspases are the final ef- fector molecules of apoptosis, responsible for the cleav- age of both cytosolic and nuclear proteins that result in the stereotypic destruction of the cell.

Extrinsic activation of apoptosis is mediated by the binding of death ligands (e.g., Fas ligand, TNF-a, TNF- related apoptosis-inducing ligand [TRAIL]) to cell sur- face transmembrane receptors (Parker and Mufti 2004;

Zang et al. 2001). For example, cytotoxic lymphocytes can initiate apoptosis by expressing Fas ligand, which binds to the Fas receptor on the surface of target cells.

This binding results in a conformational change in the cytoplasmic death domain (DD) (Chinnaiyan et al.

1995; Nagata and Golstein 1995). The altered death do- main then recruits intracellular adaptor molecules such as FADD (Fas-associated protein with a DD), which ag- gregate procaspase-8 molecules. These procaspase-8 molecules, which cleave and activate one another, in turn activate downstream caspases to induce apoptosis.

Cellular stress and damage may initiate intrinsic ag- gregation and activation of procaspases. In the best-un- derstood pathway, mitochondria are stimulated to re- lease cytochrome c into the cytosol where it forms a complex with apaf-1 (apoptotic protease-activating fac- tor-1), procaspase-9 and dATP (Li et al. 1997; Liu et al.

1996). This complex then triggers downstream effector caspases.

The Bcl-2 family of intracellular proteins includes

many of the most important regulators of apoptotic

pathways. Some proapoptotic members of this family such as Bad act by binding and inactivating death inhib- iting members of the family (Parker and Mufti 2004).

Other proapoptotic molecules such as Bax and Bak stimulate cytochrome c release from the mitochon- drion. They form homo- or heterodimers that create membrane pores/ion channels that facilitate the release of cytochrome c and other apoptogenic proteins (Des- agher et al. 1999; Hsu et al. 1997; Schendel et al. 1997;

Zha et al. 1996). Some members of this family such as Bcl-2 and Bcl-X

inhibit apoptosis by blocking release of cytochrome c. They may directly bind and sequester cytochrome c and Apaf-1 or interact with Bax or Bak, thereby inhibiting pore formation (Eskes et al. 2000;

Hu et al. 1998; Oltvai et al. 1993; Yang et al. 1997).

There are numerous other mechanisms of positive and negative regulation of apoptosis that are outside the scope of this chapter. They may involve a variety of oncogenes and tumor-suppressor genes, including p53 and c-myc.

Evidence for alterations in intramedullary apoptosis in early MDS was first suggested through morphological examination of bone marrow hematopoietic cells. In- creased apoptosis of bone marrow progenitors was pos- tulated to account for the clinical observation in early MDS of peripheral blood cytopenias in the presence of a hypercellular bone marrow. It was further postu- lated that decreased apoptosis may explain the later clinical disease progression and accumulation of imma- ture progenitor cells.

Increased apoptosis in MDS has been shown by morphology, immunohistochemistry, flow cytometry and the molecular detection of activated apoptosis-re- lated proteins (Parker et al. 2000). Raza and colleagues were the first to show increased apoptosis in patients with early MDS (Raza et al. 1995a,b). They used in situ end labeling (ISEL) of DNA strand breaks to detect apoptosis. ISEL labeling was elevated in patients with early MDS and low in AML and late MDS samples.

Using nick-end labeling (TUNEL) techniques, various groups have demonstrated a higher percentage of la- beled cells in the bone marrows of MDS patients than in healthy controls (Hellstrom-Lindberg et al. 1997; Le- pelley et al. 1996). Upregulation of Fas in MDS bone marrow samples has also been reported (Bouscary et al. 1997; Gersuk et al. 1998; Kitagawa et al. 1998). Flow cytometry studies have been utilized to better charac- terize which specific MDS marrow cells are involved in apoptosis. Using flow cytometric detection of annex-

in V on the surface of apoptotic cells, Parker and col- leagues (1998) demonstrated increased apoptosis in CD34

cells from early MDS patients compared with late MDS patients. Li et al. (2004) showed that apoptosis oc- curred predominantly (but not exclusively) in non-clo- nal cells as determined by concurrent FISH in patients with a suitable clonal marker. Rajapaksa and colleagues (1996) analyzed CD34

and CD34

subfractions of bone marrow from MDS patients and evaluated the sub-di- ploid (sub-G1) DNA peak after staining with propidium iodide. They observed that the proportion of CD34

cells with sub-G1 DNA (apoptotic) was increased in comparison to normal bone marrow and bone marrow from AML patients (Rajapaksa et al. 1996). Bcl-2 and c- Myc oncoprotein levels were also evaluated. C-Myc:Bcl-2 oncoprotein ratios were highest in early MDS samples and lower in late MDS and AML samples. The ratio of the pro-apoptotic BAX to anti-apoptotic BCL-2 was in- creased in early-stage MDS but decreased in more ad- vanced disease (Hellstrom-Lindberg et al. 1997; Parker and Mufti 2001; Parker et al. 2000). This observation supports the hypothesis that the relative balance be- tween cell-death and cell-survival signals is associated with the increased apoptosis observed in MDS progeni- tors.

The cause of abnormal apoptosis in MDS is un- known. Both intrinsic cellular defects and extrinsic fac- tors are being investigated in ongoing research. Altera- tions in the immune-mediated signals, cytokine release, and other aspects of the bone marrow microenviron- ment have been implicated.

Alterations in apoptosis appear to be a central fea- ture of MDS. Increased apoptosis is frequently observed in the early stages of MDS. Further insight into the in- trinsic and extrinsic factors that affect apoptosis is a central area of research and hold significant promise for the development of clinical therapies.

4.2.5 Immune Dysregulation

There is growing evidence that immune dysregulation

plays a role in MDS pathophysiology. The relationship

between MDS and autoimmunity stimulated the investi-

gation into the role of the immune system in MDS. The

incidence of autoimmune disorders appears to be in-

creased in patients with MDS (Saif et al. 2002). Autolo-

gous cytotoxic T lymphocytes have been observed to

exert inhibitory effects on MDS myelopoiesis in vitro.

Moreover, the features of MDS may overlap with aplastic anemia (AA) and large granular lymphocyte (LGL) dis- ease, two diseases thought to be related to autoreactive T lymphocytes (Barrett et al. 2000; Kanchan and Loughran 2003). Clinical studies indicated activity of antithymocyte globulin (ATG) and cyclosporine in the treatment of select groups of patients with MDS (Jona- sova et al. 1998; Killick et al. 2003; Molldrem et al. 2002).

Given the added observation that tumor necrosis factor alpha (TNFa) mRNA and protein levels are elevated in both bone marrow and plasma samples of patients with MDS, recent clinical trials have evaluated the efficacy of treatment with immunosuppression and anti-TNF ther- apy (Deeg et al. 2004; Kitagawa et al. 1997; Maciejewski et al. 1995; Molnar et al. 2000; Rosenfeld and Bedell 2002).

Single agent ATG has resulted in complete hemato- logic responses in up to 10±15% of patients (Killick et al. 2003). Predictors of response to immunosuppressive therapy include younger age, presence of a paroxysmal nocturnal hemoglobinuria (PNH) clone, human leuko- cyte antigen (HLA) DR15, hypocellularity, and a normal karyotype (Saunthararajah et al. 2002). Responses to ATG have been associated with disappearance of T-cell clones that demonstrate V beta clonality and which sup- press hematopoiesis ex vivo (Kochenderfer et al. 2002).

Deeg et al. treated fourteen transfusion-requiring patients with MDS with the combination of ATG and the soluble TNF receptor protein etanercept (Deeg et al. 2004). Forty-six percent of the patients responded, with five patients achieving periods of red blood cell and platelet independence that exceeded 2 years. These impressive results lend further evidence to the premise that immunomodulation may be effective in select pa- tients with MDS.

Fundamental questions remain unanswered about the precise mechanisms underlying autoimmunity in MDS. The hypothesis that T lymphocytes attack specific antigens on MDS clonal progenitors remains unproven.

Likewise, it is unclear why some patients respond to im- munosuppression and others do not. Important future investigations will include confirmation of the efficacy of immunosuppressive and anti-TNF therapies in phase III clinical trials and in identifying subsets of patients who will most benefit from these therapies.

4.2.6 Bone Marrow Microenvironment 4.2.6.1 Cytokine Milieu

The observation of increased bone marrow apoptosis in patients with early MDS stimulated the evaluation of the bone marrow microenvironment as a mediator of MDS pathophysiology. Relative deficiency or overproduction of numerous cytokines, including interleukin 1b (IL- 1b), IL-6, IL-8, stem cell factor, erythropoietin, trans- forming growth factor beta (TGF-b), GM-CSF, and TNF-a have been measured in the bone marrow and se- rum of patients with MDS with unclear and sometimes conflicting results (Bowen et al. 1993; Fontenay-Roupie et al. 1999; Maurer et al. 1993; Verhoef et al. 1992). Of all of these cytokines, increased TNF-a has been consis- tently associated with elevated Fas antigen expression on CD34

cells. Fas is a membrane protein that can ini- tiate apoptotic signals in response to crosslinking by the Fas ligand (FasL) (Gersuk et al. 1998). The resultant downstream activation of caspases, which are the criti- cal proteases that result in apoptotic cell death, indicate the importance of elevated TNF-a levels in promoting apoptosis in MDS (Mundle et al. 1999). The source of in- creased TNF-a is likely marrow macrophages and T- lymphocytes.

4.2.6.2 Neo-angiogenesis

Another aspect of the bone marrow microenvironment that has emerged as an essential factor in the pathogen- esis of MDS is neo-angiogenesis. Angiogenesis plays a critical role in tumor growth and metastasis (Folkman 1995). Increased microvessel density has been demon- strated in the bone marrow of patients with hematologic malignancies, including MDS (Alexandrakis et al. 2004, 2005; Padro et al. 2000; Pruneri et al. 1999). Neovascu- larization is mediated by a variety of angiogenic mole- cules that are released by both tumor cells and normal host cells. Abnormal elevation of several angiogenic cy- tokines and growth factors in AML samples has been re- ported. Important molecules include vascular endothe- lial growth factor (VEGF), basic fibroblast growth factor (bFGF), angiogenin, TNF-a, and TGF-b (Albitar 2001;

Alexandrakis et al. 2005; Faderl and Kantarjian 2004).

Soluble VEGF receptor has been reported as a prognos- tic factor in both AML and MDS patients (Hu et al.

2004).

4.2.7 Molecular Abnormalities Identified by Cytogenetic Features

Cytogenetic abnormalities occur in up to 70% of pa- tients with primary MDS and up to 90% of patients with therapy-related MDS (Delforge 2003). A well-described difference between primary and secondary MDS is the complexity of abnormal karyotypes. Chromosomal de- letions are common in MDS as opposed to the balanced translocations that are seen in AML. Over the last de- cade, intensive investigation has focused on identifying potential tumor suppressor genes in the regions of ge- netic loss in MDS. Here, we will review some of the in- sights based on particular chromosomal abnormalities.

4.2.7.1 Chromosome 5 Deletions

Approximately 20% of patients with MDS have abnor- malities of chromosome 5 (Third MIC Cooperative Study Group 1988). These abnormalities include inter- stitial deletions of the long arm (5q-), monosomy, and unbalanced translocations. A separate clinical entity of 5q± syndrome has been described in multiple studies in the literature and is now recognized as a separate en- tity in the WHO classification of MDS (Boultwood et al.

1994; Mathew et al. 1993). It is characterized by refrac- tory macrocytic anemia with dyserythropoiesis, a strik- ing female to male ratio of 3:1, normal or high platelet counts, 5q± as the sole cytogenetic abnormality, and a low propensity for transformation to AML. In other cases, abnormalities on chromosome 5 have been seen in familial cases of MDS and in therapy-related MDS (Grimwade et al. 1993; Mhawech and Saleem 2001).

The most critical region of deletion is presumed to lie between 5q31 and 5q33. Numerous hematopoietic growth factors are encoded on the long arm of chromo- some 5, and loss of these genes is presumed to play a role in MDS pathogenesis. The genes for IL-3, IL-4, IL-5, interferon regulator factor 1 (IRF-1), M-CSF, GM- CSF, and the receptor for M-CSF are localized on the long arm of chromosome 5. These cytokines are critical in the proliferation of granulocytes. However, it is un- clear how deletion of any one of these growth factor genes alone would result in a 5q± syndrome phenotype.

IRF-1, a transcription activator of interferon 1 genes, re- sults in a gene product that is anti-oncogenic and growth inhibitory (Harada et al. 1993). Although dele- tion of IRF-1 in one or both alleles by an accelerated exon skipping mechanism has been described in some

cases of MDS with 5q± abnormalities, specific mutations in this gene have not been described in MDS.

Other insights into important genes involved in MDS result from the analysis of genes involved in chro- mosomal translocations on chromosome 5. t(3;5)(q25.1;

q34) results in a chimeric protein, NPM-MLF1, that is associated with MDS prior to transformation to AML (Yoneda-Kato et al. 1996). t(5;11)(q31;q23) has been shown to result in a fusion between the MLL and GRAF genes in a child with JMML (Borkhardt et al. 2000).

Homologs of the human GRAF gene have putative tu- mor suppressor properties and may indicate the impor- tance of this gene in deletions of chromosome 5q.

The unique clinical activity of CC-5013 in patients with abnormalities of chromosome 5 (List et al. 2005) may afford a unique reagent with which to probe this subset of patients for specific disease-defining molecu- lar abnormalities.

4.2.7.2 Chromosome 7 Deletions

Partial or complete deletion of chromosome 7 is a com- mon finding in MDS and AML. It is seen in a variety of settings and is generally associated with poor prognosis.

Approximately 10% of 7q± deletions are seen in the set- ting of de novo MDS. The remainder of 7q± deletions are seen in cases of MDS arising after environmental or chemotherapeutic exposure and in cases related to fa- milial genetic disorders (e.g., Fanconi anemia, neurofi- bromatosis 1 [NF1], congenital neutropenia) (Mhawech and Saleem 2001). Analysis from patients with juvenile myelomonocytic leukemia, which often have mono- somy 7, has shown that approximately 30% have NF1 gene mutations (Shannon et al. 1994). NF1 functions as a tumor suppressor gene, encoding a GTPase activa- tion protein acts as a negative regulator of RAS activity (Martin et al. 1990). RAS activation occurs in a signifi- cant proportion of adult patients with MDS. Gene muta- tions of RAS or inactivation of the NF1 gene are thought to play an important role in the progression of MDS with monosomy 7 (Stephenson et al. 1995).

The region at 7q22.1 has been suggested as a critical breakpoint in myeloid malignancies (Johnson et al.

1996). Genes of interest that have been mapped to chro-

mosome 7q include erythropoietin, plasmin activator

inhibitor, T-cell receptor b, asparagine synthase gene bb

and PIK3CG.

4.2.7.3 Chromosome 20

A chromosomal 20q deletion is seen in 10% of myelo- proliferative disorders, particularly polycythemia vera, and in approximately 5% of primary MDS (Davis et al. 1984). Clinically, patients have a lower incidence of anemia and a relatively favorable prognosis. Only pa- tients with 5q± syndrome have a more prolonged life ex- pectancy. The crucial region deleted in MDS has been mapped to the region between D20S174 and D20S17 (Asimakopoulos et al. 1994; Roulston et al. 1993). Poten- tial tumor-suppressor genes of interest in this region in- clude phospholipase C d, adenosine deaminase, topo- isomerase 1, hematopoietic cell kinase, growth hormone releasing factor, myeloid leukemia gene and SRC, the human homolog of Rous sarcoma virus.

4.2.7.4 Chromosome 17

Clinically, deletion of the short arm of chromosome 17 (17p±) is seen primarily in treatment-related MDS and is characterized by dysgranulopoiesis and pseudo-Pel- ger-HuŸt anomaly (Lai et al. 1995). The p53 gene is lo- cated at 17p13.1. P53 is a critical tumor suppressor gene that has significant roles in cell cycle control, DNA re- pair, and apoptosis. Loss of p53 has been documented in a variety of cancers, and it is likely that it plays a role in a subset of MDS cases.

4.2.7.5 Trisomy 8

Trisomy 8 occurs commonly in both acute and chronic leukemias. A curious observation has been the disap- pearance of trisomy 8 clones in the course of the dis- ease. This phenomenon is independent of the percent- age of blasts in the bone marrow or the clinical status of the disease (Iwabuchi et al. 1992; Matsuda et al.

1998). As such, it is unclear how significant trisomy 8 is in the pathogenesis of MDS, and therefore, it has not attracted as much attention as some of the other cy- togenetic abnormalities.

4.2.7.6 Other Less Frequent Chromosomal Deletions

Loss of portions of chromosomes 3, 11, 12, 13, and the Y chromosome have been described with varying fre- quency in the literature. Likewise, trisomies involving chromosomes 6, 13, and 21 have been documented.

Although critical regions on each chromosome have been identified and a number of candidate tumor sup- pressor genes have been identified, the molecular pathogenesis remains elusive.

4.2.8 Translocations 4.2.8.1 TEL(ETV6) Fusion

First described by Srivastava et al. (1988) in patients with chronic myelomonocytic leukemia with eosinophi- lia, t(5;12)(q33;p13) is a recurrent chromosomal translo- cation that results in a fusion between the platelet-de- rived growth factor receptor-b (PDGRFR-b) gene on chromosome 5 and an ETS-like gene, TEL(ETV6) on chromosome 12 (Golub et al. 1994). It is now apparent that this translocation occurs in a broader group of myeloid malignancies with features of both myeloprolif- erative disorders (MPS) and MDS. ETS family members act as transcriptional activators, while PDGFR-b is a re- ceptor tyrosine kinase that acts on multiple down- stream targets including RAS. Oligomerization of the TEL/PDGRFR-b through the TEL HLH domain results in constitutive activation of the PDGRFR-b tyrosine kinase domain (Carroll et al. 1996). This constitutive ac- tivation results in cellular transformation. Demon- strated responses to the tyrosine kinase inhibitor, ima- tinib, in patients with chronic myeloproliferative dis- eases associated with TEL/PDGRFR-b lend further cre- dence to the importance of this signaling pathway in the pathogenesis of subsets of MDS/MPS (Apperley et al.

2002). TEL gene fusion has also been reported in con- junction with ARNT, MN1, EVI-1, and ACS2 in MDS pa- tients carrying a variety of translocations (Buijs et al.

1995; Raynaud et al. 1996; Salomon-Nguyen et al.

2000; Yagasaki et al. 1999).

4.2.8.2 Nucleoporin Abnormality

Nucleoporins are molecules involved in the nuclear im- port and export of proteins and RNAs (Radu et al. 1995).

The gene NUP98, in particular, has been identified as a fusion partner in patients with treatment-related AML or MDS with chromosomal translocations involving 11p15.5. Chimeric transcripts involving NUP98 combine the N-terminal GLFG repeats of NUP98 with the C-ter- minus of the partner gene (Ahuja et al. 1999; Arai et al.

1997; Nakamura et al. 1996; Nishiyama et al. 1999; Raza-

Egilmez et al. 1998; Slape and Aplan 2004). Up to 15 partnerships have been identified to date, clearly iden- tifying NUP98 as a potentially important target in ther- apy-related leukemias and MDS. The mechanism of the formation of fusion and cellular transformation is an active field of investigation.

The t(6;9)(p23;q34) translocation can occur in pa- tients with both MDS and AML. This translocation re- sults in a fusion protein between the DEK and CAN genes on chromosomes 6 and 9, respectively. CAN is structurally similar to a component of the nuclear pore complex, NUP214, which also facilitates nuclear import and export of RNA and protein (Kraemer et al. 1994).

4.2.8.3 MLL

Chromosomal translocations involving 11q23 are well described in association with biphenotypic leukemias, monocytic leukemias, infant leukemias or secondary leukemias after treatment with topoisomerase inhibitors (Ayton and Cleary 2001; Ridge and Wiedemann 1994).

The MLL gene involved in these translocations has homology to the trithorax gene of Drosophila, a homeo- tic regulator of pattern development in the fruit fly (Dja- bali et al. 1992). This suggests that MLL acts as a homeo- tic regulator of transcription.

Although the most common 11q23 translocations are involved in AML as opposed to MDS, other less com- mon MLL translocations are involved in primary or sec- ondary MDS (Mitani et al. 1995; Taki et al. 1997).

Although the conventional 11q23 translocations such as t(4;11)(q21;q23) and t(9;11)(q21;q23) have not been found in primary MDS, other translocations such as t(11;19)(q23;p13.1) and t(11;16)(q23;p13) have been re- ported. MLL gene tandem duplication has also been de- scribed in patients with MDS (Caligiuri et al. 1996).

4.2.8.4EVI-1

Alterations in chromosome 3 involving 3q21 and 3q26 bands occur in up to 2% of patients with AML or MDS (Hirai 2003), conferring a uniformly poor prog- nosis. 3q21q26 syndrome has been described in patients with abnormal megakaryocytopoiesis, elevated platelet counts, and a poor prognosis (Jotterand Bellomo et al.

1992). Genes of interest within this region include trans- ferrin, lactoferrin, transferrin receptor, melanotransfer- rin, CALLA/CD10 and ecotropic virus integration site 1

gene (EVI-1). The expression of the EVI-1 gene, located at 3q26, has generated the most interest in a variety of chromosomal abnormalities in MDS and AML. Inv(3) (q21q26) and t(3;3)(q21;q26) are classified with the 3q21q26 syndrome. The translocation t(1;3)(p36;q21) re- sults in activation of MEL1, a homologous gene to EVI-1, under the control of ribophorin I, which is in the 3q21 region (Mochizuki et al. 2000). This is associated with trilineage dysplasia and dysmegakaryocytopoiesis. The translocation t(3;21)(q26;q22) generates an AML1/EVI-1 chimeric gene that is seen in therapy-related AML/

MDS and in the blast crisis of CML (Mitani et al.

1994). The pathobiology of EVI-1-mediated leukemo- genesis is unclear, but may relate to downstream effects on TGF-b signaling (Izutsu et al. 2001).

Buonamici et al. (2004) recently generated an EVI-1 murine model of myelodysplasia. These animals devel- oped a fatal pancytopenia accompanied by a hypercellu- lar bone marrow and dyserythropoiesis.

4.2.8.5 NPM

Translocation t(3;5)(q25.1;q34) may occur in MDS and AML. It involves NPM on 5q34 and MLF1 on 3q25.1 (Yo- neda-Kato et al. 1996). NPM has a role in transporting ribosomal nucleoproteins between the nucleolus and the cytoplasm. The fusion gene may affect cell growth by altering DNA replication, RNA processing or gene ex- pression.

4.2.9 Molecular Abnormalities Unrelated to Cytogenetic Abnormalities 4.2.9.1 TP53

Mutations in the tumor suppressor, TP53, are reported in up to 10% of MDS patients (Padua et al. 1998). Studies have shown that TP53 mutations occur in predominant- ly high-risk FAB subtypes. One study showed that mu- tations resulting in loss of heterozygosity of p53 were as- sociated with therapy-related MDS and AML, deletion or loss of 5q, complex karyotype, and a uniform poor prognosis (Christiansen et al. 2001).

4.2.9.2 FLT3

Internal tandem duplication (ITD) of the FLT3 gene has

been associated with adverse outcome in AML. FLT3 is a

receptor-type tyrosine kinase that is involved in prolif- eration and differentiation in hematopoiesis. FLT3-ITD is found in up to 20% of AML and 5% of MDS (Shih et al. 2004; Yokota et al. 1997). It is associated with a high risk of transformation to AML and poor survival.

4.2.9.3 AML1

The AML1 (Runx1) gene encodes a heterodimeric tran- scription factor that is a critical regulator of hematopoi- esis. Translocations involving this gene or its binding partners present in two forms, t (8; 21) and inv(16). Mu- tations in AML1 have been identified in therapy-related MDS following alkylating agents (Christiansen et al.

2004; Harada et al. 2003). Dominant negative effects of AML1 mutations on normal AML1 function are thought to be critical in pathogenesis (Imai et al.

2000). t (3; 21) is another MDS-associated translocation involving AML1. It has been described in MDS patients exposed to organic solvents, treatment-related MDS, and the blast crisis of CML (Mitani et al. 1994; Tasaka et al. 1992).

4.2.9.4 Other Mutations

Other mutations have been described with RB, WT1, C/

EBPa and a variety of other genes with varying degrees of frequency. Their role in the pathogenesis of MDS is unclear.

4.2.10 Gene Expression Profiling

cDNA microarray technology has revolutionized molec- ular biology and medicine. For details in MDS, see Chapter 5. Using commercially available chip technol- ogy, investigators have investigated the expression of several thousands of genes in a variety of cancers, in- cluding diffuse large B cell lymphoma, follicular lym- phoma, acute myeloid leukemia and myelodysplastic syndromes (Lossos et al. 2004; Pellagatti et al. 2004;

Ross et al. 2004; Sigal et al. 2005). Identification of spe- cific molecular ªsignaturesº in these and many other cancers holds promise for further advances in predict- ing prognosis and response to therapy. The role of these genome-wide analysis tools remains to be defined. They have the potential to completely transform the way we

diagnose and prognosticate various cancers, including myelodysplasia.

4.3 Summary

The pathogenesis of MDS is complex and remains elu- sive. Figure 4.1 shows a hypothetical model. The pro- posed models agree that a multistep process occurs through which a hematopoietic stem cell is mutated and attains a growth advantage. This may occur as a re- sult of environmental damage or inherited predisposi- tion. The mutated clone is associated with morpholog- ical dysplasia, impaired differentiation and hematopoi- esis, and genomic instability. Cytokine secretion and apoptotic pathways are altered, and there may be im- pairment of immune responses. Presumably, in the early stages, increased production of proapoptotic cytokines leads to excessive apoptosis, correlating clinically with cytopenias and a cellular bone marrow. As the disease progresses, further genetic and epigenetic events occur, resulting in decreased apoptosis, clonal expansion, and progression to AML. There is considerable overlap be- tween mechanisms, and MDS is likely a heterogeneous group of diseases with a similar clinical phenotype.

Fig. 4.1. Hypothetical pathogenesis of MDS

Clinical testing of a number of molecules that affect these myriad molecular mechanisms is currently being done (Table 4.1). Characterization of genomic expres- sion patterns will inform both diagnosis and prognosti- cation. Further insight into the molecular mechanisms of MDS will provide an avenue for more tailored and ef- fective therapy in the future.

References

Abrahamson G, Boultwood J, Madden J, Kelly S, Oscier DG, Rack K, Buckle VJ, Wainscoat JS (1991) Clonality of cell populations in re- fractory anaemia using combined approach of gene loss and X- linked restriction fragment length polymorphism-methylation analyses. Br J Haematol 79:550±555

Ahuja HG, Felix CA, Aplan PD (1999) The t(11;20)(p15;q11) chromoso- mal translocation associated with therapy-related myelodysplas- tic syndrome results in an NUP98-TOP1 fusion. Blood 94:3258±

3261

Albitar M (2001) Angiogenesis in acute myeloid leukemia and myelo- dysplastic syndrome (Review). Acta Haematol 106:170±176 Alexandrakis MG, Passam FJ, Ganotakis E, Dafnis E, Dambaki C, Konso-

las J, Kyriakou DS, Stathopoulos E (2004) Bone marrow microvas- cular density and angiogenic growth factors in multiple myeloma.

Clin Chem Lab Med 42:1122±1126

Alexandrakis MG, Passam FH, Pappa CA, Sfiridaki K, Tsirakis G, Dami- lakis J, Stathopoulos EN, Kyriakou DS (2005) Relation between bone marrow angiogenesis and serum levels of angiogenin in pa- tients with myelodysplastic syndromes. Leuk Res 29:41±46 Apperley JF, Gardembas M, Melo JV, Russell-Jones R, Bain BJ, Baxter EJ,

Chase A, Chessells JM, Colombat M, Dearden CE, Dimitrijevic S, Mahon FX, Marin D, Nikolova Z, Olavarria E, Silberman S, Schultheis B, Cross NC, Goldman JM (2002) Response to imatinib mesylate in patients with chronic myeloproliferative diseases with rearrangements of the platelet-derived growth factor receptor beta. N Engl J Med 347:481±487

Arai Y, Hosoda F, Kobayashi H, Arai K, Hayashi Y, Kamada N, Kaneko Y, Ohki M (1997) The inv(11)(p15q22) chromosome translocation of de novo and therapy-related myeloid malignancies results in fu- sion of the nucleoporin gene, NUP98, with the putative RNA heli- case gene, DDX10. Blood 89:3936±3944

Asimakopoulos FA, White NJ, Nacheva E, Green AR (1994) Molecular analysis of chromosome 20q deletions associated with myelopro- liferative disorders and myelodysplastic syndromes. Blood 84:3086±3094

Attwood JT, Yung RL, Richardson BC (2002) DNA methylation and the regulation of gene transcription (Review). Cell Mol Life Sci 59:241±

257

Ayton PM, Cleary ML (2001) Molecular mechanisms of leukemogenesis mediated by MLL fusion proteins [Review]. Oncogene 20:5695±

Backx B, Broeders L, Touw I, Lowenberg B (1993) Blast colony-forming 5707 cells in myelodysplastic syndrome: decreased potential to gener- ate erythroid precursors. Leukemia 7:75±79

Backx B, Broeders L, Hoefsloot LH, Wognum B, Lowenberg B (1996) Ery- thropoiesis in myelodysplastic syndrome: expression of receptors for erythropoietin and kit ligand. Leukemia 10:466±472 Barrett J, Saunthararajah Y, Molldrem J (2000) Myelodysplastic syn-

drome and aplastic anemia: distinct entities or diseases linked by a common pathophysiology? (Review). Semin Hematol 37:15±29

Bellamy WT, Richter L, Sirjani D, Roxas C, Glinsmann-Gibson B, Frutiger Y, Grogan TM, List AF (2001) Vascular endothelial cell growth fac- tor is an autocrine promoter of abnormal localized immature mye- loid precursors and leukemia progenitor formation in myelodys- plastic syndromes. Blood 97:1427±1434

Table 4.1. Examples of therapies currently under clini- cal evaluation for MDS and presumed their mechanism of action

Agents Mechanism of action

TLK199 Glutathione analog

Arsenic trioxide Pro-apoptotic, antiangio- genesis, differentiation

Etanercept Soluble TNF receptor

Pentoxifylline Anti-TNF agent

Infliximab Anti-TNF antibody

CEP701 FLT-3 inhibitor

Imatinib Tyrosine kinase inhibitor Tipifarnib Farnesyl transferase inhibitor Lonafarnib Farnesyl transferase inhibitor 5-azacitidine DNA methyltransferase

inhibitor 2'-deoxy-5-azacitidine

(decitabine)

DNA methyltransferase inhibitor

Sodium phenylbutyrate Histone deacetylase inhibitor MS-275 Histone deacetylase inhibitor Valproic acid Histone deacetylase inhibitor Suberoylanilide hy-

droxamic acid (SAHA)

Histone deacetylase inhibitor

FK228 (Depsipeptide) Histone deacetylase inhibitor Thalidomide Antiangiogenesis?,

immunomodulatory Lenalidomide Antiangiogenesis?,

immunomodulatory

PTK787 VEGF receptor kinase inhibitor

Bevacizumab Anti-VEGF antibody

AG3340 Matrix metalloproteinase

inhibitor

Bortezomib Proteasome inhibitor

SCIO 469 MAP Kinase inhibitor

Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DAG, Gralnick HR, Sultan C, The French-American-British (FAB) Co-Operative Group (1982) Proposals for the classification of the myelodysplas- tic syndromes. Br J Haematol 51:189±199

Borkhardt A, Bojesen S, Haas OA, Fuchs U, Bartelheimer D, Loncarevic IF, Bohle RM, Harbott J, Repp R, Jaeger U, Viehmann S, Henn T, Korth P, Scharr D, Lampert F (2000) The human GRAF gene is fused to MLL in a unique t(5;11)(q31;q23) and both alleles are disrupted in three cases of myelodysplastic syndrome/acute myeloid leuke- mia with a deletion 5q. Proc Natl Acad Sci USA 97:9168±9173 Boultwood J, Lewis S, Wainscoat JS (1994) The 5q±syndrome (Review).

Blood 84:3253±3260

Bouscary D, De Vos J, Guesnu M, Jondeau K, Viguier F, Melle J, Picard F, Dreyfus F, Fontenay-Roupie M (1997) Fas/Apo-1 (CD95) expression and apoptosis in patients with myelodysplastic syndromes. Leu- kemia 11:839±845

Bowen D, Yancik S, Bennett L, Culligan D, Resser K (1993) Serum stem cell factor concentration in patients with myelodysplastic syn- dromes. Br J Haematol 85:63±66

Broxmeyer HE, Cooper S, Li ZH, Lu L, Song HY, Kwon BS, Warren RE, Donner DB (1995) Myeloid progenitor cell regulatory effects of vascular endothelial cell growth factor. Int J Hematol 62:203±215 Buijs A, Sherr S, van Baal S, van Bezouw S, van der Plas D, Geurts vK, Riegman P, Lekanne DR, Zwarthoff E, Hagemeijer A (1995) Trans- location (12;22) (p13;q11) in myeloproliferative disorders results in fusion of the ETS-like TEL gene on 12p13 to the MN1 gene on 22q11 [erratum appears in Oncogene 1995 11:809]. Oncogene 10:1511±1519

Buonamici S, Li D, Chi Y, Zhao R, Wang X, Brace L, Ni H, Saunthararajah Y, Nucifora G (2004) EVI1 induces myelodysplastic syndrome in mice. J Clin Invest 114:713±719

Caligiuri MA, Strout MP, Schichman SA, Mrozek K, Arthur DC, Herzig GP, Baer MR, Schiffer CA, Heinonen K, Knuutila S, Nousiainen T, Ruutu T, Block AW, Schulman P, Pedersen-Bjergaard J, Croce CM, Bloom- field CD (1996) Partial tandem duplication of ALL1 as a recurrent molecular defect in acute myeloid leukemia with trisomy 11. Can- cer Res 56:1418±1425

Cameron EE, Bachman KE, Myohanen S, Herman JG, Baylin SB (1999) Synergy of demethylation and histone deacetylase inhibition in the re-expression of genes silenced in cancer. Nat Genet 21:103±107

Carroll M, Tomasson MH, Barker GF, Golub TR, Gilliland DG (1996) The TEL/platelet-derived growth factor beta receptor (PDGF beta R) fusion in chronic myelomonocytic leukemia is a transforming pro- tein that self-associates and activates PDGF beta R kinase-depen- dent signaling pathways. Proc Natl Acad Sci USA 93:14845±14850 Chen H, Sandler DP, Taylor JA, Shore DL, Liu E, Bloomfield CD, Bell DA (1996) Increased risk for myelodysplastic syndromes in individuals with glutathione transferase theta 1 (GSTT1) gene defect. Lancet 347:295±297

Chinnaiyan AM, O'Rourke K, Tewari M, Dixit VM (1995) FADD, a novel death domain-containing protein, interacts with the death do- main of Fas and initiates apoptosis. Cell 81:505±512

Christiansen DH, Andersen MK, Pedersen-Bjergaard J (2001) Mutations with loss of heterozygosity of p53 are common in therapy-related myelodysplasia and acute myeloid leukemia after exposure to al- kylating agents and significantly associated with deletion or loss

of 5q, a complex karyotype, and a poor prognosis. J Clin Oncol 19:1405±1413

Christiansen DH, Andersen MK, Pedersen-Bjergaard J (2004) Mutations of AML1 are common in therapy-related myelodysplasia following therapy with alkylating agents and are significantly associated with deletion or loss of chromosome arm 7q and with subsequent leukemic transformation. Blood 104:1474±1481

Daskalakis M, Nguyen TT, Nguyen C, Guldberg P, Kohler G, Wijermans P, Jones PA, Lubbert M (2002) Demethylation of a hypermethylated P15/INK4B gene in patients with myelodysplastic syndrome by 5- Aza-2'-deoxycytidine (decitabine) treatment. Blood 100:2957±

2964

Davis MP, Dewald GW, Pierre RV, Hoagland HC (1984) Hematologic manifestations associated with deletions of the long arm of chro- mosome 20. Cancer Genet Cytogenet 12:63±71

Deeg HJ, Jiang PYZ, Holmberg LA, Scott B, Petersdorf EW, Appelbaum FR (2004) Hematologic responses of patients with MDS to anti- thymocyte globulin plus etanercept correlate with improved flow scores of marrow cells. Leuk Res 28:1177±1180

Delforge M (2003) Understanding the pathogenesis of myelodysplastic syndromes (Review). Hematol J 4:303±309

Desagher S, Osen-Sand A, Nichols A, Eskes R, Montessuit S, Lauper S, Maundrell K, Antonsson B, Martinou JC (1999) Bid-induced confor- mational change of Bax is responsible for mitochondrial cyto- chrome c release during apoptosis. J Cell Biol 144:891±901 Djabali M, Selleri L, Parry P, Bower M, Young BD, Evans GA (1992) A

trithorax-like gene is interrupted by chromosome 11q23 translo- cations in acute leukaemias [erratum appears in Nat Genet 1993 4:431]. Nat Genet 2:113±118

Engelhardt M, Wasch R, Guo Y (2004) Telomeres and telomerase in nor- mal and leukemic hematopoietic cells. Leuk Res 28:1001±1004 Eskes R, Desagher S, Antonsson B, Martinou JC (2000) Bid induces the

oligomerization and insertion of Bax into the outer mitochondrial membrane. Mol Cell Biol 20:929±935

Esteller M, Herman JG (2002) Cancer as an epigenetic disease: DNA methylation and chromatin alterations in human tumours (Re- view). J Pathol 196:1±7

Estey EH (2004) Modulation of angiogenesis in patients with myelo- dysplastic syndrome. Bailliere's Best Practice in Clinical Haematol- ogy 17:623±639

Faderl S, Kantarjian HM (2004) Novel therapies for myelodysplastic syn- dromes (Review). Cancer 101:226±241

Farquhar MJ, Bowen DT (2003) Oxidative stress and the myelodysplas- tic syndromes (Review). Int J Hematol 77:342±350

Fischle W, Wang Y, Allis CD (2003) Histone and chromatin cross-talk (Review). Current Opinion in Cell Biology 15:172±183

Folkman J (1995) Angiogenesis in cancer, vascular, rheumatoid and other disease (Review). Nat Med 1:27±31

Fontenay-Roupie M, Bouscary D, Guesnu M, Picard F, Melle J, Lacombe C, Gisselbrecht S, Mayeux P, Dreyfus F (1999) Ineffective erythro- poiesis in myelodysplastic syndromes: correlation with Fas expres- sion but not with lack of erythropoietin receptor signal transduc- tion. Br J Haematol 106:464±473

Fuhler GM, Drayer AL, Vellenga E (2003) Decreased phosphorylation of

protein kinase B and extracellular signal-regulated kinase in neu-

trophils from patients with myelodysplasia. Blood 101:1172±1180

Garfinkel L, Boffetta P (1990) Association between smoking and leu- kemia in two American Cancer Society prospective studies. Can- cer 65:2356±2360

Gaymes TJ, North PS, Brady N, Hickson ID, Mufti GJ, Rassool FV (2002) Increased error-prone non homologous DNA end-joining ± a pro- posed mechanism of chromosomal instability in Bloom's syn- drome. Oncogene 21:2525±2533

Gersuk GM, Beckham C, Loken MR, Kiener P, Anderson JE, Farrand A, Troutt AB, Ledbetter JA, Deeg HJ (1998) A role for tumor necrosis factor- a, Fas and Fas-Ligand in marrow failure associated with myelodysplastic syndrome. Br J Haematol 103:176±188 Golub TR, Barker GF, Lovett M, Gilliland DG (1994) Fusion of platelet-

derived growth factor receptor b to a novel Ets-like gene, tel, in chronic myelomonocytic leukemia with t(5;12) chromosomal translocation. Cell 77:307±316

Greenberg PL (1998) Apoptosis and its role in the myelodysplastic syn- dromes: implications for disease natural history and treatment (Review). Leuk Res 22:1123±1136

Grimwade DJ, Stephenson J, De Silva C, Dalton RG, Mufti GJ (1993) Familial MDS with 5q± abnormality. Br J Haematol 84:536±538 Hake SB, Xiao A, Allis CD (2004) Linking the epigenetic 'language' of

covalent histone modifications to cancer (Review). Br J Cancer 90:761±769

Harada H, Kitagawa M, Tanaka N, Yamamoto H, Harada K, Ishihara M, Taniguchi T (1993) Anti-oncogenic and oncogenic potentials of interferon regulatory factors-1 and -2. Science 259:971±974 Harada H, Harada Y, Tanaka H, Kimura A, Inaba T (2003) Implications of

somatic mutations in the AML1 gene in radiation-associated and therapy-related myelodysplastic syndrome/acute myeloid leuke- mia. Blood 101:673±680

Harris NL, Jaffe ES, Diebold J, Flandrin G, Muller-Hermelink HK, Vardi- man J, Lister TA, Bloomfield CD (1999) World Health Organization classification of neoplastic diseases of the hematopoietic and lymphoid tissues: report of the clinical advisory committee meet- ing, Airlie House, Virginia, November 1997. J Clin Oncol 17:3835±

3849

Hellstrom-Lindberg E, Kanter-Lewensohn L, Ost A (1997) Morphologi- cal changes and apoptosis in bone marrow from patients with myelodysplastic syndromes treated with granulocyte-CSF and er- ythropoietin. Leuk Res 21:415±425

Henikoff S, Matzke MA (1997) Exploring and explaining epigenetic ef- fects. Trends Genet 13:293±295

Herman JG, Baylin SB (2003) Gene silencing in cancer in association with promoter hypermethylation (Review). N Engl J Med 349:2042±2054

Hirai H (2003) Molecular mechanisms of myelodysplastic syndrome (Review). Jpn J Clin Oncol 33:153±160

Hoefsloot LH, van Amelsvoort MP, Broeders LC, van dPD, van Lom K, Hoogerbrugge H, Touw IP, Lowenberg B (1997) Erythropoietin-in- duced activation of STAT5 is impaired in the myelodysplastic syn- drome. Blood 89:1690±1700

Hofmann WK, Kalina U, Koschmieder S, Seipelt G, Hoelzer D, Ottmann OG (2000) Defective megakaryocytic development in myelodys- plastic syndromes (Review). Leuk Lymphoma 38:13±19 Hsu YT, Wolter KG, Youle RJ (1997) Cytosol-to-membrane redistribution

of Bax and Bcl-X(L) during apoptosis. Proc Natl Acad Sci USA 94:3668±3672

Hu Y, Benedict MA, Wu D, Inohara N, Nunez G (1998) Bcl-XL interacts with Apaf-1 and inhibits Apaf-1-dependent caspase-9 activation.

Proc Natl Acad Sci USA 95:4386±4391

Hu Q, Dey AL, Yang Y, Shen Y, Jilani IB, Estey EH, Kantarjian HM, Giles FJ, Albitar M (2004) Soluble vascular endothelial growth factor recep- tor 1, and not receptor 2, is an independent prognostic factor in acute myeloid leukemia and myelodysplastic syndromes. Cancer 100:1884±1891

Imai Y, Kurokawa M, Izutsu K, Hangaishi A, Takeuchi K, Maki K, Ogawa S, Chiba S, Mitani K, Hirai H (2000) Mutations of the AML1 gene in myelodysplastic syndrome and their functional implications in leukemogenesis. Blood 96:3154±3160

Issa J-PJ, Garcia-Manero G, Giles FJ, Mannari R, Thomas D, Faderl S, Bayar E, Lyons J, Rosenfeld CS, Cortes J, Kantarjian HM (2004) Phase 1 study of low-dose prolonged exposure schedules of the hypomethylating agent 5-aza-2'-deoxycytidine (decitabine) in hematopoietic malignancies. Blood 103:1635±1640

Iwabuchi A, Ohyashiki K, Ohyashiki JH, Sasao I, Murakami T, Kodama A, Toyama K (1992) Trisomy of chromosome 8 in myelodysplastic syndrome. Significance of the fluctuating trisomy 8 population.

Cancer Genet Cytogenet 62:70±74

Izutsu K, Kurokawa M, Imai Y, Maki K, Mitani K, Hirai H (2001) The cor- epressor CtBP interacts with Evi-1 to repress transforming growth factor beta signaling. Blood 97:28152±822

Johnson EJ, Scherer SW, Osborne L, Tsui LC, Oscier D, Mould S, Cotter FE (1996) Molecular definition of a narrow interval at 7q22.1 asso- ciated with myelodysplasia. Blood 87:3579±3586

Jonasova A, Neuwirtova R, Cermak J, Vozobulova V, Mocikova K, Sisko- va M, Hochova I (1998) Cyclosporin A therapy in hypoplastic MDS patients and certain refractory anaemias without hypoplastic bone marrow. Br J Haematol 100:304±309

Jotterand Bellomo M, Parlier V, Muhlematter D, Grob JP, Beris P (1992) Three new cases of chromosome 3 rearrangement in bands q21 and q26 with abnormal thrombopoiesis bring further evidence to the existence of a 3q21q26 syndrome (Review). Cancer Genet Cy- togenet 59:138±160

Kalina U, Hofmann WK, Koschmieder S, Wagner S, Kauschat D, Hoelzer D, Ottmann OG (2000) Alteration of c-mpl-mediated signal trans- duction in CD34(+) cells from patients with myelodysplastic syn- dromes. Exp Hematol 28:1158±1163

Kanchan K, Loughran TP (2003) Antigen-driven clonal T cell expansion in disorders of hematopoiesis. Leuk Res 27:291±292

Kantarjian HM, Cortes JE, Rowe JM, Kurzrock R (2002) Future clinical implications for farnesyltransferase inhibitors in hematologic ma- lignancies (Review). Semin Hematol 39:36±38

Killick SB, Mufti G, Cavenagh JD, Mijovic A, Peacock JL, Gordon-Smith EC, Bowen DT, Marsh JC (2003) A pilot study of antithymocyte glo- bulin (ATG) in the treatment of patients with `low-risk' myelodys- plasia. Br J Haematol 120:679±684

Kitagawa M, Saito I, Kuwata T, Yoshida S, Yamaguchi S, Takahashi M, Tanizawa T, Kamiyama R, Hirokawa K (1997) Overexpression of tu- mor necrosis factor (TNF)-alpha and interferon (IFN)-gamma by bone marrow cells from patients with myelodysplastic syndromes.

Leukemia 11:2049±2054

Kitagawa M, Yamaguchi S, Takahashi M, Tanizawa T, Hirokawa K, Ka-

miyama R (1998) Localization of Fas and Fas ligand in bone mar-

row cells demonstrating myelodysplasia. Leukemia 12:486±492

Kochenderfer JN, Kobayashi S, Wieder ED, Su C, Molldrem JJ (2002) Loss of T-lymphocyte clonal dominance in patients with myelo- dysplastic syndrome responsive to immunosuppression. Blood 100:3639±3645

Kraemer D, Wozniak RW, Blobel G, Radu A (1994) The human CAN pro- tein, a putative oncogene product associated with myeloid leuke- mogenesis, is a nuclear pore complex protein that faces the cyto- plasm. Proc Natl Acad Sci USA 91:1519±1523

Kurzrock R, Fenaux P, Raza A, Mufti G, Aul C, Germing U, Kerstens R, DePorre P (2004) High-risk myelodysplastic syndrome (MDS): first results of international phase 2 study with oral farnesyltransferase inhibitor R115777 (ZARNESTRA A ) [Abstract]. Blood 104:23 a

Lai JL, Preudhomme C, Zandecki M, Flactif M, Vanrumbeke M, Lepelley

P, Wattel E, Fenaux P (1995) Myelodysplastic syndromes and acute myeloid leukemia with 17p deletion. An entity characterized by specific dysgranulopoiesis and a high incidence of P53 mutations.

Leukemia 9:370±381

Langland G, Elliott J, Li Y, Creaney J, Dixon K, Groden J (2002) The BLM helicase is necessary for normal DNA double-strand break repair.

Cancer Res 62:2766±2770

Lepelley P, Campergue L, Grardel N, Preudhomme C, Cosson A, Fenaux P (1996) Is apoptosis a massive process in myelodysplastic syn- dromes? Br J Haematol 95:368±371

Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, Alnemri ES, Wang X (1997) Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cas- cade. Cell 91:479±489

Li X, Bryant E, Deeg HJ (2004) Simultaneous demonstration of clonal chromosome abnormalities and apoptosis in individual marrow cells in myelodysplastic syndrome. Int J Hematol 80:140±145 List A, Kurtin S, Roe DJ, Buresh A, Mahadevan D, Fuchs D, Rimsza L,

Heaton R, Knight R, Zeldis JB (2005) Efficacy of lenalidomide in myelodysplastic syndromes. N Engl J Med 352:549±557 Liu X, Kim CN, Yang J, Jemmerson R, Wang X (1996) Induction of apop-

totic program in cell-free extracts: requirement for dATP and cy- tochrome c. Cell 86:147±157

Lossos IS, Czerwinski DK, Alizadeh AA, Wechser MA, Tibshirani R, Bot- stein D, Levy R (2004) Prediction of survival in diffuse large-B-cell lymphoma based on the expression of six genes. N Engl J Med 350:1828±1837

Lubbert M (2003) Gene silencing of the p15/INK4B cell-cycle inhibitor by hypermethylation: an early or later epigenetic alteration in myelodysplastic syndromes? (Review). Leukemia 17:1762±1764 Maciejewski J, Selleri C, Anderson S, Young NS (1995) Fas antigen ex-

pression on CD34+ human marrow cells is induced by interferon gamma and tumor necrosis factor alpha and potentiates cyto- kine-mediated hematopoietic suppression in vitro. Blood 85:3183±3190

Martin GA, Viskochil D, Bollag G, McCabe PC, Crosier WJ, Haubruck H, Conroy L, Clark R, O'Connell P, Cawthon RM (1990) The GAP-re- lated domain of the neurofibromatosis type 1 gene product inter- acts with ras p21. Cell 63:843±849

Mathew P, Tefferi A, Dewald GW, Goldberg SL, Su J, Hoagland HC, Noel P (1993) The 5q± syndrome: a single-institution study of 43 con- secutive patients. Blood 81:1040±1045

Matsuda A, Yagasaki F, Jinnai I, Kusumoto S, Murohashi I, Bessho M, Hirashima K (1998) Trisomy 8 may not be related to the pathogen- esis of myelodysplastic syndromes: disappearance of trisomy 8 in

a patient with refractory anaemia without haematological im- provement. Eur J Haematol 60:260±261

Maurer AB, Ganser A, Buhl R, Seipelt G, Ottmann OG, Mentzel U, Geiss- ler RG, Hoelzer D (1993) Restoration of impaired cytokine secretion from monocytes of patients with myelodysplastic syndromes after in vivo treatment with GM-CSF or IL-3. Leukemia 7:1728±1733 Mayani H, Baines P, Bowen DT, Jacobs A (1989) In vitro growth of mye-

loid and erythroid progenitor cells from myelodysplastic patients in response to recombinant human granulocyte-macrophage col- ony-stimulating factor. Leukemia 3:29±32

Mhawech P, Saleem A (2001) Myelodysplastic syndrome: review of the cytogenetic and molecular data (Review). Critical Reviews in On- cology-Hematology 40:229±238

Mitani K, Ogawa S, Tanaka T, Miyoshi H, Kurokawa M, Mano H, Yazaki Y, Ohki M, Hirai H (1994) Generation of the AML1-EVI-1 fusion gene in the t(3;21)(q26;q22) causes blastic crisis in chronic myelocytic leukemia. EMBO J 13:504±510

Mitani K, Kanda Y, Ogawa S, Tanaka T, Inazawa J, Yazaki Y, Hirai H (1995) Cloning of several species of MLL/MEN chimeric cDNAs in myeloid leukemia with t(11;19)(q23;p13.1) translocation. Blood 85:2017±

2024

Mochizuki N, Shimizu S, Nagasawa T, Tanaka H, Taniwaki M, Yokota J, Morishita K (2000) A novel gene, MEL1, mapped to 1p36.3 is highly homologous to the MDS1/EVI1 gene and is transcription- ally activated in t(1;3)(p36;q21)-positive leukemia cells. Blood 96:3209±3214

Molldrem JJ, Leifer E, Bahceci E, Saunthararajah Y, Rivera M, Dunbar C, Liu J, Nakamura R, Young NS, Barrett AJ (2002) Antithymocyte glo- bulin for treatment of the bone marrow failure associated with myelodysplastic syndrome [summary for patients in Ann Intern Med 2002 137:I-27]. Ann Intern Med 137:156±163

Molnar L, Berki T, Hussain A, Nemeth P, Losonczy H (2000) Detection of TNFalpha expression in the bone marrow and determination of TNFalpha production of peripheral blood mononuclear cells in myelodysplastic syndrome. Pathology Oncology Research 6:18±

23

Mundle SD, Reza S, Ali A, Mativi Y, Shetty V, Venugopal P, Gregory SA, Raza A (1999) Correlation of tumor necrosis factor alpha (TNF al- pha) with high Caspase 3-like activity in myelodysplastic syn- dromes. Cancer Lett 140:201±207

Nagata S, Golstein P (1995) The Fas death factor. Science 267:14491456 Nakamura T, Largaespada DA, Lee MP, Johnson LA, Ohyashiki K, Toya- ma K, Chen SJ, Willman CL, Chen IM, Feinberg AP, Jenkins NA, Co- peland NG, Shaughnessy JD (1996) Fusion of the nucleoporin gene NUP98 to HOXA9 by the chromosome translocation t(7;11)(p15;p15) in human myeloid leukaemia. Nat Genet 12:154±158

Nakayama M, Wada M, Harada T, Nagayama J, Kusaba H, Ohshima K, Kozuru M, Komatsu H, Ueda R, Kuwano M (1998) Hypomethyla- tion status of CpG sites at the promoter region and overexpres- sion of the human MDR1 gene in acute myeloid leukemias. Blood 92:4296±4307

Nilsson L, Astrand-Grundstrom I, Anderson K, Arvidsson I, Hokland P,

Bryder D, Kjeldsen L, Johansson B, Hellstrom-Lindberg E, Hast R,

Jacobsen SE (2002) Involvement and functional impairment of

the CD34

CD38

Thy-1

hematopoietic stem cell pool in myelo-

dysplastic syndromes with trisomy 8. Blood 100:259±267