Histone Deacetylase Inhibitors for the Treatment of Myelodysplastic Syndrome and Acute Myeloid Leukemia

Histone deacetylase inhibitors for the treatment of myelodysplastic syndrome and acute myeloid leukemia

A Quinta´s-Cardama1,3, FPS Santos1,2,3 and G Garcia-Manero1

1Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, TX, USA and 2Hematology Department, Hospital Israelita Albert Einstein, Sa˜o Paulo, Brazil

Epigenetic changes have been identiﬁed in recent years as drugs whose mechanism of action has been linked to modula- important factors in the pathogenesis of myelodysplastic tion of gene transcription have taken center stage of both syndrome (MDS) and acute myeloid leukemia (AML). Histone clinical and preclinical research in AML and MDS.8 deacetylase inhibitors (HDACIs) regulate the acetylation of histones as well as other non-histone protein targets. Treat- Histone deacetylase inhibitors (HDACIs) represent one class ment with HDACIs results in chromatin remodeling that permits of gene expression modulating drugs, which is currently being re-expression of silenced tumor suppressor genes in cancer developed for therapy of several types of cancer.9 HDACIs were cells, which, in turn, can potentially result in cellular differ- initially developed as putative epigenetic agents, owing to their entiation, inhibition of proliferation and/or apoptosis. Several ability to modify acetylation status of histones and induce gene classes of HDACIs are currently under development for the expression.10 Typically, optimal gene expression occurs when treatment of patients with MDS and AML. Although modest 11 clinical activity has been reported with the use of HDACIs as histones are in a state of maximal acetylation. Therefore, single-agent therapy, marked responses have been observed in HDACIs have the potential to cause re-expression of genes selected subsets of patients. More importantly, HDACIs appear abnormally suppressed in cancer cells, thus potentially reversing to be synergistic in vitro and improve response rates in vivo the malignant phenotype and inducing growth inhibition, cell- when combined with other agents, such as hypomethylating cycle arrest, differentiation and/or apoptosis.9 However, more agents. Furthermore, HDACIs are also being investigated in recent results of laboratory and clinical studies have indicated combination with non-epigenetic therapies. This article synthe- sizes the most recent results reported with HDACIs in clinical that besides modulation of gene expression, HDACIs have a trials conducted in patients with MDS and other myeloid diverse array of mechanisms of action, and the complexity of malignancies. their effects on malignant cells has only begun to be unraveled.9 Leukemia (2011) 25, 226–235; doi:10.1038/leu.2010.276; HDACIs are being actively investigated as potential therapy of published online 30 November 2010 MDS and AML. Herein, we summarize the results of the most Keywords: histone deacetylase inhibitors; myelodysplastic important clinical and preclinical studies with these drugs.12 syndromes; acute myeloid leukemia

Histone modiﬁcations and transcriptional activity

Introduction DNA is wrapped around the nucleosome, an octamer composed of two units of each of the main histones (H2A, H2B, H3 and H4), Myelodysplastic syndromes (MDS) are a group of clonal and forms a ‘beads-on-a-string’ structure, which folds into hematopoietic stem cell disorders characterized by ineffective higher-order chromatin.13 Chromatin structure modifications are hematopoiesis, peripheral blood cytopenias and a propensity to 1 important modulators of the transcriptional activity of specific transform to acute myeloid leukemia (AML). AML is a myeloid genes. Post-transcriptional modifications of histones have an malignancy characterized by increased self-renewal, limited 2 important role in changing the chromatin structure and thus differentiation and deregulated proliferation of myeloid blasts. modulating gene expression.13 Covalent histone modifications Treatment outcomes of both MDS and AML are suboptimal. include acetylation, methylation, phosphorylation, poly(ADP)- Outside the context of stem cell transplantation, MDS remains ribosylation and ubiquitination.13 Certain modifications of an incurable malignancy, and most patients with AML treated specific amino-acid residues in histones lead to a more open with chemotherapeutic regimens relapse and perish to their 3,4 chromatin configuration (euchromatin), which makes it more disease or associated complications. AML and MDS are accessible to the transcription machinery, whereas other typically diagnosed in elderly individuals (median age 65–70 modifications change the chromatin into a more tightly wound years), and an intensive treatment approach can be applied only structure (heterochromatin), which is associated with transcrip- in the minority of patients owing to increased therapy-related 14,15 5–7 tional silencing. It is believed that the combination of mortality and decreased efficacy. Thus, there is a great need specific histone modifications constitutes a ‘histone code’ that for new agents with more benign toxicity profiles and improved determines the state of the chromatin and thus the transcrip- activity for the treatment of MDS and AML. In recent years, tional activity of genomic regions.14 Acetylation is one of the main histone modifications Correspondence: Dr A Quinta´s-Cardama, Department of Leukemia, associated with the regulation of gene expression.15 Histones Unit 428, The University of Texas MD Anderson Cancer Center, 1515 possess N-terminal tails that are rich in lysine. Acetylation of Holcombe Boulevard, Houston, TX 77030, USA. lysine residues opens the chromatin and promotes gene E-mail: [email protected] 3These authors contributed equally to this work. transcription, whereas histone deacetylation promotes chroma- 15 Received 25 March 2010; revised 30 September 2010; accepted 15 tin condensation and represses gene transcription. Histone October 2010; published online 30 November 2010 deacetylation is mediated by histone deacetylases (HDACs).15 HDAC inhibitors for MDS and AML A Quinta´s-Cardama et al 227 HDACs are conventionally divided into families: class I, including the human HDACs 1, 2, 3 and 8; class II, which encompasses the human HDACs 5, 6, 7, 9 and 10; class III includes the NAD-dependent HDACs related to the yeast Sir2 protein (sirtuins); and class IV, defined by the recently identified human HDAC 11.15,16 Class I, II and IV HDACs are zinc (Zn2 þ )- dependent enzymes and are considered the classical HDACs, distinctive from the unrelated sirtuins.17 Available HDACIs only target the classical HDACs, whereas sirtuin inhibitors are currently under development. Besides histones, HDACs also have several other substrates, including transcription factors (for example, p53, signal transducer and activator of transcription 1 and 3), signaling mediators, steroid receptors, chaperone proteins (heat-shock protein 90 (Hsp90)) and DNA repair enzymes (Ku70).18 Thus, a more appropriate name for these enzymes would be ‘lysine deacetylases’.18 Figure 1 Chemical structure of representative compounds from the major classes of HDACIs. Short-chain fatty acids (valproic acid), Histone deacetylase inhibitors hydroxamic derivatives (vorinostat), non-hydroxamate small mole- cule inhibitors (MGCD0103) and cyclic peptides (depsipeptide (romidepsin)). Increased activity of HDACs leading to reduced or abnormal histone acetylation patterns has been described in cancer cells.19,20 It has been shown that malignant cells present with higher level of HDACs than normal tissues, generating factors and other proteins involved in transcription.18 Micro- hypoacetylated histones.20 Aberrant HDAC activity has also array analyses of leukemic cell lines after exposure to HDACIs been described in certain leukemia subtypes that carry chimeric (trichostatin, vorinostat, romidepsin) have revealed modulation proteins resulting from translocations involving transcription of a large number of genes (approximately 10–20% of the factors such as core binding factor (t(8;21)(q22;q22)FAML1-ETO; genome) repressing and inducing the expression of genes in inv(16)FCBFA-MYH11) and retinoic acid receptor alpha equal measure.26,27 One of the more commonly HDACI- (RARA; t(15;17)(q24;q21)FPML-RARA).21–24 These chimeric induced genes is the cyclin-dependent kinase (CDK) inhibitor proteins recruit HDACs to co-repressor complexes and block CDKN1A (p21WAF1/CIP1).28–30 Induction of CDKN1A by vorino- transcription of genes that lead to myeloid differentiation. Use of stat is p53 independent and accompanied by changes in all-trans retinoic acid (ATRA) in patients with PML-RARA- the acetylation pattern of histones in the CDKN1A gene positive acute promyelocytic leukemia relieves the transcrip- promoter.28,30 Increase in the level of p21WAF1/CIP1 is linked to tional blockade and impedes recruitment of HDACs, thus cell-cycle arrest which is observed upon exposure to HDACIs, leading to cellular differentiation.25 but is not essential for HDACI-induced apoptosis.31,32 HDACIs The development of HDACIs started with the discovery that also decrease the expression of cyclin proteins and CDKs several compounds that were known to induce differentiation and lead to hypophosphorylation of retinoblastoma protein, of leukemic cell lines were inhibitors of HDACs and induced contributing to cell-cycle arrest.33 Mitotic arrest and mitotic cell hyperacetylated histones.10 This led to the development of death in tumor cells is observed after exposure to HDACIs. other compounds with HDAC-inhibitory activity, based on the HDACIs increase acetylation of histone H3K9 in pericentro- hypothesis that inhibition of HDACs would induce histone meric chromatin, which interferes with assembly of the hyperacetylation and restore gene expression in cancer cells. kinetochore.34 This leads to mitotic arrest at prometaphase, This could potentially lead to cell differentiation, inhibition of followed by aberrant mitosis with chromosome segregation proliferation and/or apoptosis. Many structurally diverse com- defects and eventually apoptosis/cell death.35,36 pounds can bind to HDACs and inhibit their enzymatic activity, HDACIs can lead to cell death by activating apoptosis by both including hydroxamates, cyclic peptides, aliphatic acids, the intrinsic and the extrinsic pathways. HDACIs increase levels benzamides and electrophilic ketones (Figure 1). They differ in of the death receptor ligands FasL and TRAIL (tumor necrosis their potency and specificity against different classes of HDACs factor-related apoptosis-inducing ligand), as well as of death (Table 1). receptors Fas and DR5 in leukemic cells, indicating activation of the extrinsic apoptotic pathway.32,37 Upregulation of death receptors and their ligands is not observed in normal hemato- Mechanisms of action of HDACI poietic cells, indicating selectivity of these compounds for malignant cells.32 Increased TRAIL expression is owing to HDACIs exert a myriad of biological effects in cancer cells, increased acetylation of histones located at the TRAIL gene including induction of differentiation/apoptosis, cell-cycle/mitotic promoter and increased acetylation of the transcription factors arrest and induction of autophagic cell death (Figure 2).18 Normal Sp1 and Sp3, which stimulate the expression of TRAIL.32 cells seem to be resistant to cell death induced by HDACIs, and Interestingly, blocking TRAIL expression did not abrogate this selectivity could be exploited in the therapy of cancer apoptosis in response to HDACIs in leukemic cells, indicating patients. The mechanisms of action of HDACIs are complex, and that additional mechanisms of apoptosis are responsible for cell to what extent the clinical activity of HDACIs is due to changes in death observed with these compounds.32 Consistent with this transcription or effects on non-histone proteins is currently contention, several studies have shown that HDACIs activate the unknown and subject of much investigation. intrinsic apoptotic pathway dependent on the mitochondria. HDACI can modify gene expression secondary to acetylation HDACIs have been shown to increase levels of pro-apoptotic of histones and by altering the acetylation status of transcription Bcl-2 family proteins (Bax, Bak, Bim, Bmf),38–40 to downregulate

Leukemia HDAC inhibitors for MDS and AML A Quinta´s-Cardama et al 228 Table 1 Chemical classes of histone deacetylase inhibitors

Chemical class Agent Inhibitory range In vitro effect Phase of development

SCFA Phenylbutyrate mM Apoptosis, cell-cycle arrest, differentiation I/II VPA mM Apoptosis, differentiation II Hydroxamic acid Vorinostat mM Apoptosis, cell-cycle arrest, differentiation I/II Panobinostat nM Apoptosis, cell-cycle arrest, differentiation I/II LAQ824 nM Apoptosis, cell-cycle arrest, differentiation I PXD101 nM Apoptosis, cell-cycle arrest, differentiation I/II Trichostatin A nM Apoptosis, cell-cycle arrest, differentiation Preclinical Oxamﬂatin mM Apoptosis, cell-cycle arrest Preclinical Scriptaid mM Apoptosis, cell-cycle arrest, differentiation Preclinical Pyroxamide nM Apoptosis, cell-cycle arrest, differentiation Preclinical CBHA mM Apoptosis, cell-cycle arrest, differentiation Preclinical Cyclic tetrapeptides CHAP nM Cell-cycle arrest Preclinical Trapoxin A nM Cell-cycle arrest, differentiation Preclinical Depsipeptide (FK228) mM Apoptosis, cell-cycle arrest I/II Apicidin nM Cell-cycle arrest, apoptosis Preclinical Sulfonamide anilide MGCD-0103 nM Apoptosis, cell-cycle arrest I/II Benzamide Entinostat mM Cell-cycle arrest I/II Epoxide Depudecin mM Cell-cycle arrest, differentiation Preclinical Abbreviations: CBHA, m-carboxy cinnamic acid bishydroxamic acid; CHAP, cyclic hydroxamic acid-containing peptide (trapoxin analog); SCFA, short-chain fatty acids; VPA, valproic acid.

breaks. HDACIs have been shown to decrease levels of non-homologous end-joining proteins Ku80 and Rad50 and to decrease DNA binding of Ku70 by increasing its acetylation.50,51 Production of ROS may be an important mechanism in cell-death induced by HDACIs.31,41 Several HDACI (vorinostat, entinostat, LAQ-824) have been shown to increase intracellular ROS levels,31,41,43 and HDACI’s cytotoxicity is blocked by pre-exposure to antioxidants.43 Exposure to the antioxidant N-acetylcysteine blocks mitochondrial injury and apoptosis induced by HDACIs in leukemic cells, suggesting that production of ROS is probably connected to activation of the mitochondrial death pathway.31,41 One recent report shows that vorinostat increases activity of the ROS-generating enzyme NADPH oxidase (NOX) in leukemic cell lines HL60, U937 and ML1, indicating that NOX is one of the major sources of increased ROS levels in vorinostat-treated cells.52 Importantly, Figure 2 Potential mechanisms of action of HDACIs (see text for details). levels of the endogenous antioxidant thioredoxin are increased in response to HDACIs in normal cells compared with transformed cells, and no increase in ROS levels is observed antiapoptotic proteins (Bcl-2, Bcl-XL, MCL-1, XIAP, survi- in normal cells after exposure to vorinostat.53 In contrast, vin)38,40,41 and to enhance Bid cleavage and its subsequent HDACIs increase the expression of thioredoxin binding protein, activation.42,43 Exposure to HDACIs leads to increased cyto- a negative regulator of thioredoxin, in transformed cells, and this plasmic levels of cytochrome C and Smac, indicating mitochon- contributes to the observed increase in ROS levels.53 Increase in drial injury and loss of mitochondrial membrane potential ROS levels is followed by activation of an antioxidant response, (Dcm).41,43 Increased expression of antiapoptotic Bcl-2 family which is one of the mechanisms of resistance to HDACIs.52 proteins, such as Bcl-2, Bcl-XL and XIAP, has been associated Exposure of leukemic cells U937 to vorinostat leads to trans- with resistance to HDACI-induced apoptosis.41,44 location of the transcription factor nuclear factor erythroid HDACIs can cause DNA damage. HDACIs are known to 2-like 2 (also known as Nrf2) to the nucleus, in which it sensitize cells to radiation and DNA-damaging chemotherapeu- transactivates its target genes.52 Nrf2 is a master regulator of the tic agents,45–47 and preclinical studies have shown that HDACIs antioxidant response, and it increases expression of several induce phosphorylation of H2AX (g-H2AX) and ATM, which are antioxidant genes, including GST (glutathione S-transferase), markers of DNA damage.48,49 g-H2AX foci were located in the GSR (glutathione reductase), GCLC (glutathione synthase), same chromatin strands as acetylated H3 and H4 histones.48 SOD1 and SOD2 (superoxide dismutase 1 and 2), in response Induction of DNA damage was followed by apoptosis with to vorinostat.52 In a phase I clinical trial of vorinostat cond- cleavage of caspase 3 and poly(ADP-ribose) polymerase.48 ucted in patients with AML/MDS, increased expression of 17 Generation of g-H2AX foci in response to the HDACI LAQ-824 antioxidant genes (including TXN (thioredoxin), SOD1 and depends on increased generation of reactive oxygen species SOD2) correlated with resistance to vorinostat.54 In addition, (ROS) (see below).50 HDACI also contribute to DNA damage by leukemic cells resistant to pan-HDACIs (HL-60/LR cells) display impairing mechanisms of DNA repair, such as the non-homo- decreased production of ROS after exposure to the HDACI LAQ- logous end-joining pathway for repair of DNA double-strand 824 in comparison to HDACI-sensitive cells.44 After exposure to

Leukemia HDAC inhibitors for MDS and AML A Quinta´s-Cardama et al 229 HDACIs, leukemic cells depend on the glutathione system to established. There is rationale for utilizing HDACIs in combination buffer increased levels of ROS, and depleting intracellular with other agents, such as hypomethylating drugs and conventional glutathione might be exploited to increase cellular sensitivity chemotherapeutic agents. Several in vitro studies have revealed that to HDACIs.52 Combined exposure of vorinostat and the HDACIs and DNA hypomethylating agents act synergistically to glutathione depleting compound phenylethyl isothiocyanate induce re-expression of silenced genes in cancer cells.60 Several substantially increased the production of ROS and cytotoxicity trials are currently exploring the clinical utility of such approach. in leukemic cells following exposure to vorinostat.52 HDACI could be combined with other agents that generate ROS and cause DNA damage, such as anthracyclines (see below). In Valproic acid aggregate, these results suggest that HDACI-mediated apoptosis Valproic acid (VPA, 2-propylpentanoic acid) is a short-chain is, in part, dependent on the generation of ROS, with subsequent fatty acid oral anticonvulsant that has been shown to induce DNA and mitochondrial damage and activation of the intrinsic differentiation of transformed hematopoietic progenitor cells apoptotic pathway. and leukemic blasts from bone marrow and peripheral blood of Inhibition of caspases does not completely prevent cell death patients with AML.61 Besides its HDAC inhibitory activity, VPA induced by HDACIs such as vorinostat.55 Caspase-independent also synergizes with ATRA in the differentiation induction of cell death in response to HDACIs has morphological features of AML blasts in vitro.62,63 A phase I study of single-agent VPA autophagic cell death.55 Autophagy is a physiological cell death rendered responses in 44% of patients with high-risk MDS and process characterized by the degradation of intracellular AML.64 In a follow-up report from the same group, 75 patients organelles by the lysosomal machinery.56 Combination of with MDS (N ¼ 43) or AML (N ¼ 32) were treated with VPA, HDACIs (MGCD0103 or vorinostat) with the BH3-mimetic including nine patients who also received ATRA from the start of compound GX15-070 induces both apoptosis and autophagy in treatment.65 Hematologic improvement (HI) was observed in leukemic cell lines and blasts from patients with AML, and 24% of patients, with higher response rates in those with MDS inhibiting autophagy with chloroquine significantly increased and normal blast counts (52%). The addition of ATRA did not cell viability, indicating that autophagy is an important improve response rates. Other studies using VPA either as single mechanism of cell death in response to these drugs.57 The agent or in combination with ATRA as therapy of poor-risk ability of HDACIs to induce autophagy and caspase-indepen- AML/MDS have also reported minimal activity.66–69 Thus, it dent cell death opens the door to the use of these compounds in seems that the activity of single-agent VPA in MDS is mostly the treatment of neoplastic cells bearing apoptotic defects. limited to patients with low-risk MDS and normal blast counts. Another potential mechanism of action of HDACI is blocking A phase I/II study evaluated the combination of VPA the activity of the chaperone protein Hsp90.58 Chaperone (escalating doses starting at 50 mg/kg orally daily for 7 days), proteins assist in the folding of client proteins and prevent their 5-azacitidine (75 mg/m2 for 7 days) and ATRA (45 mg/m2 for 5 aggregation into non-functional structures and degradation by days, starting on day 3) in 53 patients with AML/MDS (relapsed/ the proteasome. Hsp90 client proteins include several pro- refractory or untreated patients over 60 years of age).70 The survival proteins such as Bcr-Abl, FLT3, EGFR, VEFGR and overall response rate (ORR) was 42%, including 12 patients Raf1.58 Activity of Hsp90 is dependent on its acetylation status, (22%) with complete remissions (CR). The ORR in the subgroup with hyperacetylation abrogating Hsp90 function.58,59 The main of patients without previous therapy was 52%. The maximum- HDAC responsible for removing acetyl residues from Hsp90 tolerated dose (MTD) of VPA was 50 mg/kg, and DLTs is the cytoplasmic HDAC6. Inhibition of HDAC6 leads were somnolence and confusion. Higher levels of VPA were to hyperacetylation and decreased activity of Hsp90, with found in responders compared with non-responders (132 versus consequent polyubiquitination and proteasomal degradation of 104 mg/ml; Po0.005). However, there was no correlation its client proteins in leukemic cells.58 Hsp90 antagonists between VPA levels and histone acetylation or between histone (17-N-allylamino-17-demethoxy geldanamycin, 17-dimethyl- acetylation and clinical responses. aminoethylamino-17-demethoxy-geldanamycin) are currently VPA has also been combined with decitabine in two clinical being developed and investigated in clinical trials of cancer studies. In the first report, 54 patients (AML ¼ 48; MDS ¼ 6) were patients, and there is potential for combining these compounds treated with decitabine (15 mg/m2 intravenously (i.v.) daily for 10 with HDACIs.44 Development of resistance to HDACI is associated days) along with escalating doses of VPA (starting at 20 mg/kg) for with increased expression of HDAC 1, 2 and 4, but decreased 10 days.71 The MTD of VPA was 50 mg/kg. The ORR was 22% expression of HDAC6.44 This leads to increased acetylation of (CR ¼ 19%; CR with incomplete platelet recovery ¼ 3%). Histone Hsp90, and these cells show collateral sensitivity to Hsp90 acetylation was detected in 35% of patients receiving VPA at the inhibitors in vitro.44 Thus, therapy with Hsp90 inhibitors might higher dose of 50 mg/kg. Transient DNA hypomethylation was potentially over-ride resistance to HDAC inhibitors.44 observed and associated with reactivation of the CDKN2B gene. A preliminary analysis of an ongoing randomized phase II study evaluated the addition of VPA (50 mg/kg for 5 days) to therapy with Results of clinical trials with HDACIs decitabine (20 mg/m2 i.v. for 5 days).72 Seventy-four patients (AML ¼ 23; MDS ¼ 43; chronic myelomonocytic leukemia ¼ 8) Several clinical trials are being conducted with HDACIs in patients were treated. The addition of VPA did not improve the ORR (52% with MDS and/or AML (Table 2). It is worth noting that most of the (combination) versus 43% (single-agent decitabine), P ¼ non- heretofore available information on HDACIs derives from phase I significant (NS)) or time to response (57 days (combination) versus trials and is, therefore, preliminary. In general, single-agent HDACI 64 days (single agent), P ¼ NS). therapy has been associated with low response rates, usually in the range of 10–20%. A common pattern of toxicity has been observed, with fatigue, nausea, vomiting and diarrhea being the most Vorinostat (suberoylanilide hydroxamic acid; SAHA; common dose-limiting toxicities (DLTs) with most HDACIs. Zolinza) Increases in histone acetylation have been detected by most Vorinostat is a potent HDACI containing a hydroxamic acid studies, but no correlation with clinical responses has been moiety that binds to the zinc-containing pocket in the catalytic

Leukemia HDAC inhibitors for MDS and AML A Quinta´s-Cardama et al 230 Table 2 Clinical outcomes in selected studies of HDACI in patients with MDS and/or AML

Agent Study and No. of Treatment RR, no. (%)a reference patients

VPA Kuendgen 75 VPA dose to target a serum concentration of 50–100 mg/ml+ATRA 18 (24) et al.65 80 mg/m2/day orally Â 7 days every 2 weeks Bug et al.66 26 VPA starting at 5–10 mg/kg daily, increased 5 mg/kg every 3–4 days 2 (10) until 50 mg/kg or toxicity+ATRA 45 mg/m2 daily Kuendgen 58 VPA dose to target serum concentration of 50–100 mg/ml+ATRA 3 (5) et al.67 80 mg/m2/day orally Â 7 days every 2 weeks Raffoux et al.68 11 VPA dose to target serum concentration of 50–100 mg/ml+ATRA 3 (27) 45 mg/m2/day orally starting 1 week after VPA+theophylline dose to target a serum concentration of 10–15 mg/ml Pilatrino et al.69 20 VPA dose to target serum concentration of 45–100 mg/ml+ATRA 6 (30) 45 mg/m2/day starting after VPA reached target concentration Soriano et al.70 53 5-azacitidine 75 mg/m2/day subcutaneous for 7 days+VPA 50–75 22 (42) mg/kg orally for 7 days+ATRA 45 mg/m2/day orally for 5 days (starting on day 3) Garcia-Manero 54 Decitabine 15 mg/m2/day i.v. for 10 days+VPA 20–50 mg/kg/day orally 12 (22) et al.71 for 10 days every 4 weeks Issa et al.72 76 Decitabine 20 mg/m2/day i.v. for 5 days±VPA 50 mg/kg/day orally for 14 (52) 7 days vs 17 (43) Vorinostat Garcia-Manero 41 100–300 mg orally 2–3 times daily for 14 days 7 (17) et al.54 Schaefer et al.77 37 Vorinostat 400 mg daily for 21 days (arm A); vorinostat 200 mg three 1 (4.5) times daily for 14 days (arm B) Silverman et al.78 28 5-azacitidine 55–75 mg/m2/day subcutaneous for 7 days+vorinostat 18 (86) 200–300 mg orally two times daily for 3–14 days Ravandi et al.80 30 Decitabine 10–25 mg/m2/day i.v. for 5 days+vorinostat 100–200 mg 5 (16) orally 3 times daily for 14 days Kirschbaum 72 Decitabine 20 mg/m2 i.v. for 5 days+vorinostat 400 mg orally once daily 41% et al.79 (concomitantly or sequentially to decitabine) for 7–14 days (concurrent); 21% (sequentially) Garcia-Manero 52 Vorinostat 500 mg three times daily for 3 days, followed by idarubicin 36 (80)b et al.82 12 mg/m2/day i.v. for 3 days and cytarabine 1.5 g/m2/day i.v. for 4 days MGCD0103 Garcia-Manero 29 20–80 mg/m2 orally three times a week Â 3 weeks 3 (10) et al.84 Garcia-Manero 52 5-azacitidine 75 mg/m2/day subcutaneous for 7 days+MGCD0103 11 (30) et al.85 35–110 mg orally three times a week Entinostat Gojo et al.86 39 4–8 mg/m2 orally once a week Â 2–4 every 4–6 weeks 0 (0) Gore et al.88 31 5-azacitidine 30–50 mg/m2/day subcutaneous for 10 days+entinostat 12 (44) 2–8 mg/m2 orally on days 3 and 10 Depsipeptide Byrd et al.95 10 13 mg/m2 i.v. once a week Â 3 weeks 0 (0) Panobinostat Giles et al.91 29 4.8–14 mg/m2 i.v. once daily Â 7 every 3 weeks 0 (0) Abbreviations: AML, acute myeloid leukemia; ATRA, all-trans retinoic acid; HDACI, histone deacetylase inhibitors; i.v., intravenously; MDS, myelodysplastic syndrome; no., number; RR, response rate; VPA, valproic acid. aResponse criteria varied according to the study. bAmong 45 evaluable patients.

site of HDAC1, 2, 3 and 6, causing their reversible inhibition.73 (17%) patients (all with AML) had a response: two CR, two CRi Vorinostat is the ﬁrst HDACI compound that was approved for (CR with incomplete blood count recovery) and three HI. All the treatment of a malignant disease (primary cutaneous T-cell patients who responded were receiving less than the MTD lymphoma).74 Initial phase I studies of vorinostat in patients (100 mg/m2 three times daily, 200 mg/m2 three times daily and with lymphoma and solid tumors showed that the prolonged use 200 mg/m2 twice daily). Histone H3 acetylation was observed in of this drug was limited by the development of anorexia, all patients at all dose levels tested. Gene expression analysis vomiting, diarrhea, dehydration and fatigue, which typically revealed that the upregulation of ROS scavenger genes (see occurred after 2 weeks of therapy and were reversible upon above) was associated with resistance to vorinostat, whereas drug discontinuation.75,76 In a phase I study for patients with downregulation of proliferation-associated genes was observed relapsed/refractory MDS or leukemia (AML ¼ 31; MDS ¼ 3; in patients who had clinical improvement.54 If validated in a chronic lymphocytic leukemia ¼ 4; chronic myeloid leuk- prospective manner, the predictive value of these gene emia ¼ 1; acute lymphoblastic leukemia ¼ 2), vorinostat was expression signatures may allow for exclusion of non-responders administered at two dose schedules: orally three times daily for from future clinical trials. In another study, 37 patients with AML 14 days every 21 days or orally twice daily for 14 days every 21 were treated with vorinostat in a phase II trial, but only one days.54 The MTD was established at either 250 mg three times patient had a clinical response.77 daily or 200 mg twice daily for 14 days every 21 days. DLTs Vorinostat was combined with 5-azacitidine in a phase I were fatigue, nausea, vomiting and diarrhea. Overall, seven study.78 Patients (MDS ¼ 20; AML ¼ 8) received different doses

Leukemia HDAC inhibitors for MDS and AML A Quinta´s-Cardama et al 231 and schedules of 5-azacitidine (55–75 mg/m2/day for 7 days) 3 and 11.83 In a dose escalation study, MGCD0103 was and vorinostat (200–300 mg orally, 2–3 times daily for 3–14 administered three times weekly to 29 patients with relapsed/ days). The ORR was 86% (CR 43%), and median time to refractory leukemia or MDS and older patients with untreated response was two cycles. Main non-hematological toxicities AML or MDS.84 The MTD was 60 mg/m2, with DLTs (fatigue, were grade 1–2 fatigue and anorexia, which were more frequent vomiting and diarrhea) observed at the dose level of 80 mg/m2. with longer duration of vorinostat therapy (14 days). Vorinostat Dose-dependent histone acetylation occurred in all patients, was also combined with decitabine in two independent phase I and a maximal effect was seen at the 60 mg/m2 dose level. trials. In one study, 72 patients with AML/MDS received Pharmacokinetics showed that drug exposure increased with vorinostat (400 mg once daily, concomitant or following subsequent doses up to 60 mg/m2. Two patients with relapsed/ decitabine) and decitabine (20 mg/m2 i.v. daily for 5 days).79 refractory AML and one patient with MDS achieved a marrow The ORR (CR þ PR þ HI) was 41% in the concomitant schedule CR (less than 5% blasts). and 21% in the sequential schedule. Toxicity occurred in 89% The combination of 5-azacitidine (75 mg/m2/day for 7 days) of patients (most frequently grade 1–2 nausea, vomiting, fatigue, and MGCD0103 (escalating doses starting at 35 mg daily three constipation and diarrhea) being severe in 24% of them. In times weekly, starting on day 5) was evaluated in a phase I/II another phase I trial, 31 patients with refractory/relapsed trial in 57 patients with relapsed/refractory AML/MDS.85 The leukemia (including 21 with AML and three with MDS) received ORR was 30% (11 out of 30 evaluable patients). The MTD of decitabine at escalating doses (10–25 mg/m2 i.v. daily for 5 MGCD0103 in this trial was 90 mg three times weekly and days) followed by vorinostat (100–200 mg orally three times observed DLTs included nausea, vomiting, diarrhea, anorexia daily for 2 weeks).80 Only one patient achieved CR, whereas and dehydration. A reduction in HDAC activity in peripheral four others had reduction in bone marrow blasts. Adverse events blood mononuclear cells was observed in most treated patients. included pulmonary embolism, diarrhea, neutropenia, fatigue, syncope and renal failure. The combination of vorinostat and Entinostat (MS-275) hypomethylating agents (5-azacitidine or decitabine) appears Entinostat is a benzamide HDACI that has been shown to induce active in high-risk MDS and AML. However, the ideal dose and differentiation and apoptosis of human leukemia cells in vivo.31 schedule of vorinostat to be used in combination with these In a phase I trial, entinostat was orally administered to 39 adults drugs still needs to be determined. with advanced acute leukemia (AML ¼ 37; chronic myeloid Preliminary evidence suggests that HDACIs are amenable to leukemia-blast crisis ¼ 1; acute undifferentiated leukemia ¼ 1).86 combination with standard chemotherapeutic agents. The The MTD was defined as 8 mg/m2 weekly for 4 weeks every 6 activity of idarubicin has been evaluated in combination with weeks. DLTs included neurological toxicity and laboratory vorinostat on the leukemia cell lines MOLT4 and HL60.49 The abnormalities (hypertriglyceridemia and hyperglycemia). One combination of idarubicin with vorinostat at 0.075 mM resulted patient had a CR in the bone marrow (blasts decreased from 54 in at least an additive loss of viability in both cell lines. The to 3% in the first cycle) and three other patients had partial combination of these agents increased histone H3 and H4 response in the bone marrow with temporary decreases in blasts acetylation at 24 h, phosphorylated H2AX and induced counts, but no responses by AML International Working Group CDKN2A mRNA. In another study, the leukemia cell lines response criteria were observed.87 Entinostat induced histone HL60 and K562 were exposed to vorinostat, cytarabine and H3/H4 acetylation, CDKN1A expression and caspase-3 activa- etoposide.81 Concomitant use of vorinostat and cytarabine led tion in bone marrow mononuclear cells. to cytotoxic antagonism, whereas giving vorinostat first followed A combined study of 5-azacitidine and entinostat was by a drug-free interval before cytarabine produced synergism. conducted in patients with AML (N ¼ 14), MDS (N ¼ 13) and Etoposide given after vorinostat was also synergistic. Analyses of chronic myelomonocytic leukemia (N ¼ 4).88 Patients received cell cycle revealed that vorinostat induces cell-cycle arrest in entinostat (2–8 mg/m2 on days 3 and 10) and 5-azacitidine G1 or G2 phase, with subsequent recovery into the S phase after (30–50 mg/m2 subcutaneous on days 1–10). Responses were treatment is stopped, explaining the sequence-dependent seen in 44% of patients, including 7.5% CR. DLTs were synergism. Investigators at MD Anderson Cancer Center recently observed with entinostat at 8 mg/m2 (laryngeal edema, pro- reported on the results of a phase II trial combining vorinostat longed neutropenia, fatigue). Fatigue was common with with idarubicin and cytarabine in patients with AML.82 Fifty-two entinostat doses of 6 mg/m2, which precluded prolonged patients with newly diagnosed AML received vorinostat (500 mg therapy. DNA methylation analysis of four different tumor three times daily for 3 days) followed on day 4 by idarubicin suppressor genes (CDKN2B, CDH1, DAPK1 and SOCS1) (12 mg/m2/day for 3 days) and cytarabine (1.5 g/m2/day by conducted during the first cycle of therapy revealed that reversal continuous infusion for 4 days). Responding patients received of DNA promoter methylation was observed both in responders five cycles of consolidation with the same dose of vorinostat, but and non-responders. Similarly, increases in histone H3/H4 reduced doses of idarubicin and cytarabine, followed by acetylation were observed in 92% of patients, with no maintenance therapy with vorinostat (200 mg three times daily correlation to clinical responses.89 for 14 days, every 4 weeks, for 12 cycles). The CR rate was 80%, and no unexpected toxicities were observed. The median disease-free survival was 9.5 months and median overall Panobinostat (LBH589) survival has not been reached yet, but follow-up is short. The Panobinostat is a cinnamic hydroxamic acid analog that has potent study is currently ongoing and a randomized trial is warranted to activity as a HDACI and has been shown to induce apoptosis in better define the value of adding vorinostat to standard AML cell lines.90 Panobinostat was evaluated in a phase I study induction chemotherapy in AML. involving 15 patients (AML ¼ 13; MDS ¼ 1; acute lymphoblastic leukemia ¼ 1).91 Patients received panobinostat as a 30-min i.v. infusion on days 1 to 7 every 21 days. Reversible grade 3 QTc MGCD0103 prolongation was observed in four patients at 11.5 mg/m2. The MGCD0103 is an orally bioavailable sulfonamide anilide median acetylation of histones H2B and H3 in CD34 þ blasts and derivative isotype-specific HDACI, which targets isoforms 1, 2, B cells significantly increased on therapy as did apoptosis in

Leukemia HDAC inhibitors for MDS and AML A Quinta´s-Cardama et al 232 CD14 þ cells.91 No clinical responses were observed other than Conﬂict of interest transient reductions in peripheral blood blasts. The authors declare no conﬂict of interest.

Romidepsin (FK228; depsipeptide) Author contributions Romidepsin is a cyclic tetrapeptide derivative that is converted to its active conformation after being reduced intracellularly.92 93 AQC and FPSS designed the research and wrote the manuscript. Romidepsin selectively inhibits HDAC isotypes 1, 2, 4 and 6. GGM supervised the research and approved the manuscript. Romidepsin was recently approved for treatment of primary cutaneous T-cell lymphoma.94 In a phase I trial, romidepsin was administered at a dose of 13 mg/m2 on days 1, 8 and 15 of a 28- day cycle to patients with AML (n ¼ 10) and chronic lympho- References cytic leukemia (n ¼ 10).95 There was increased H3 histone acetylation at 4 h in all five patients analyzed, but no clinical 1 Heaney ML, Golde DW. Myelodysplasia. N Engl J Med 1999; 340: responses were achieved. Development of progressive fatigue 1649–1660. and nausea precluded repeated treatment. 2 Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H et al. World Health Organization Classification of Tumours of Haema- topoietic and Lymphoid Tissues, 4th edn. IARC Press: Lyon, 2008. 3 Ravandi F, Burnett AK, Agura ED, Kantarjian HM. Progress in Other HDACIs the treatment of acute myeloid leukemia. Cancer 2007; 110: Several more specific and potent HDACI are undergoing 1900–1910. different stages of preclinical and/or clinical development 4 Estey E, Thall P, Beran M, Kantarjian H, Pierce S, Keating M. Effect of diagnosis (refractory anemia with excess blasts, refractory (Table 1). For many of these novel agents, clinical data are anemia with excess blasts in transformation, or acute myeloid either unavailable or still very preliminary. Similar to panobino- leukemia [AML]) on outcome of AML-type chemotherapy. Blood stat, LAQ824 and PXD101 are hydroxamic acid derivatives with 1997; 90: 2969–2977. potent in vitro and in vivo antineoplastic activity that are 5 Kantarjian H, Ravandi F, O’Brien S, Cortes J, Faderl S, Garcia- currently being explored in clinical studies.96,97 Manero G et al. Intensive chemotherapy does not benefit most older patients (age 70 years or older) with acute myeloid leukemia. Blood 2010, (doi: 10.1182/blood-2010-03-276485). 6 Estey E, de Lima M, Tibes R, Pierce S, Kantarjian H, Champlin R Future perspectives et al. Prospective feasibility analysis of reduced-intensity con- ditioning (RIC) regimens for hematopoietic stem cell transplanta- Although promising, the precise role of HDACIs in the therapy tion (HSCT) in elderly patients with acute myeloid leukemia (AML) and high-risk myelodysplastic syndrome (MDS). Blood 2007; 109: of MDS/AML remains to be defined. Overall, the use of HDACIs 1395–1400. used as single agents in patients with these malignancies yields 7 Kantarjian H, O’Brien S, Cortes J, Giles F, Faderl S, Jabbour E et al. modest results. Yet, selected patients appear to derive significant Results of intensive chemotherapy in 998 patients age 65 years or benefit from these compounds. Indeed, some patients achieve older with acute myeloid leukemia or high-risk myelodysplastic CR after having failed several lines of therapy and/or having syndrome: predictive prognostic models for outcome. Cancer become refractory to conventional chemotherapeutic agents. 2006; 106: 1090–1098. 8 Griffiths EA, Gore SD. DNA methyltransferase and histone On the basis of available data, it appears that combinations of deacetylase inhibitors in the treatment of myelodysplastic syn- HDACIs with active agents in MDS such as DNA methylation dromes. Semin Hematol 2008; 45: 23–30. inhibitors (that is, azacitidine or decitabine) or even with 9 Minucci S, Pelicci PG. Histone deacetylase inhibitors and the conventional cytotoxic chemotherapy may enhance the anti- promise of epigenetic (and more) treatments for cancer. Nat Rev leukemic activity of the latter. To further improve the clinical Cancer 2006; 6: 38–51. activity of HDACIs, two critical aspects of HDACI therapy 10 Richon VM, Emiliani S, Verdin E, Webb Y, Breslow R, Rifkind RA et al. A class of hybrid polar inducers of transformed cell warrant further investigation: (1) the exact mechanisms by differentiation inhibits histone deacetylases. Proc Natl Acad Sci which HDACIs work remain unclear. Although gene expression USA 1998; 95: 3003–3007. modulation is generally accepted as an important mechanism of 11 Marks P, Rifkind RA, Richon VM, Breslow R, Miller T, Kelly WK. action of HDACIs, a large variety of transcription factors and Histone deacetylases and cancer: causes and therapies. Nat Rev transcriptional co-regulators are modified by acetylation. Cancer 2001; 1: 194–202. HDACIs have been shown to induce CDKN1A expression, a 12 Bhalla KN. Epigenetic and chromatin modifiers as targeted therapy of hematologic malignancies. J Clin Oncol 2005; 23: 3971–3993. key mediator of the tumor suppressor activity of p53 and appear 13 Felsenfeld G, Groudine M. Controlling the double helix. Nature to also alter the regulation of the tumor suppressor activity of 2003; 421: 448–453. the retinoblastoma protein. A better understanding of the in vivo 14 Rice JC, Allis CD. Code of silence. Nature 2001; 414: 258–261. mechanisms of action of HDACIs against leukemic cells will 15 Yang XJ, Seto E. HATs and HDACs: from structure, function and facilitate the development of combination strategies with regulation to novel strategies for therapy and prevention. novel agents, which, like Hsp90 antagonists or CDK inhibitors, Oncogene 2007; 26: 5310–5318. 16 Brandl A, Heinzel T, Kramer OH. Histone deacetylases: salesmen hold great promise for the treatment of MDS and AML; and customers in the post-translational modification market. Biol (2) clinical evidence indicates that not all patients will benefit Cell 2009; 101: 193–205. from HDACI therapy. It will therefore be important to identify 17 Gregoretti I, Lee Y-M, Goodson HV. Molecular evolution of the reliable biomarkers such as specific gene signatures associated histone deacetylase family: functional implication of phylogenetic with response to better discern which patients will benefit the analysis. J Mol Biol 2004; 338: 17–31. most from HDACI therapy. The enrollment of patients in clinical 18 Xu WS, Parmigiani RB, Marks PA. Histone deacetylase inhibitors: molecular mechanisms of action. Oncogene 2007; 26: trials of HDACIs will certainly aid to answer these questions 5541–5552. and to define the potential of these agents for the treatment of 19 Fraga MF, Ballestar E, Villar-Garea A, Boix-Chornet M, Espada J, MDS and AML. Schotta G et al. Loss of acetylation at Lys16 and trimethylation at

Leukemia HDAC inhibitors for MDS and AML A Quinta´s-Cardama et al 233 Lys20 of histone H4 is a common hallmark of human cancer. induction through activation of proapoptotic protein Bim. Nat Genet 2005; 37: 391–400. Proc Natl Acad Sci USA 2005; 102: 16090–16095. 20 Nakagawa M, Oda Y, Eguchi T, Aishima S, Yao T, Hosoi F et al. 40 Xu W, Ngo L, Perez G, Dokmanovic M, Marks PA. Intrinsic Expression profile of class I histone deacetylases in human cancer apoptotic and thioredoxin pathways in human prostate cancer cell tissues. Oncol Rep 2007; 18: 769–774. response to histone deacetylase inhibitor. Proc Natl Acad Sci USA 21 Lin RJ, Nagy L, Inoue S, Shao W, Miller Jr WH, Evans RM. Role 2006; 103: 15540–15545. of the histone deacetylase complex in acute promyelocytic 41 Rosato RR, Maggio SC, Almenara JA, Payne SG, Atadja P, Spiegel S leukaemia. Nature 1998; 391: 811–814. et al. The histone deacetylase inhibitor LAQ824 induces human 22 Gelmetti V, Zhang J, Fanelli M, Minucci S, Pelicci PG, Lazar MA. leukemia cell death through a process involving XIAP down- Aberrant recruitment of the nuclear receptor corepressor–histone regulation, oxidative injury, and the acid sphingomyelinase- deacetylase complex by the acute myeloid leukemia fusion partner dependent generation of ceramide. Mol Pharmacol 2006; 69: ETO. Mol Cell Biol 1998; 18: 7185–7191. 216–225. 23 Grignani F, De Matteis S, Nervi C, Tomassoni L, Gelmetti V, 42 Rosato RR, Almenara JA, Dai Y, Grant S. Simultaneous activation Cioce M et al. Fusion proteins of the retinoic acid receptor–alpha of the intrinsic and extrinsic pathways by histone deacetylase recruit histone deacetylase in promyelocytic leukaemia. (HDAC) inhibitors and tumor necrosis factor-related apoptosis- Nature 1998; 391: 815–818. inducing ligand (TRAIL) synergistically induces mitochondrial 24 Martens JH, Brinkman AB, Simmer F, Francoijs KJ, Nebbioso A, damage and apoptosis in human leukemia cells. Mol Cancer Ther Ferrara F et al. PML-RARalpha/RXR alters the epigenetic landscape 2003; 2: 1273–1284. in acute promyelocytic leukemia. Cancer Cell 2010; 17: 173–185. 43 Ruefli AA, Ausserlechner MJ, Bernhard D, Sutton VR, Tainton KM, 25 Altucci L, Gronemeyer H. The promise of retinoids to fight against Kofler R et al. The histone deacetylase inhibitor and chemother- cancer. Nat Rev Cancer 2001; 1: 181–193. apeutic agent suberoylanilide hydroxamic acid (SAHA) induces a 26 Chambers AE, Banerjee S, Chaplin T, Dunne J, Debernardi S, cell-death pathway characterized by cleavage of Bid and produc- Joel SP et al. Histone acetylation-mediated regulation of genes in tion of reactive oxygen species. Proc Natl Acad Sci USA 2001; 98: leukaemic cells. Eur J Cancer 2003; 39: 1165–1175. 10833–10838. 27 Peart MJ, Smyth GK, van Laar RK, Bowtell DD, Richon VM, 44 Fiskus W, Rao R, Fernandez P, Herger B, Yang Y, Chen J et al. Marks PA et al. Identification and functional significance of genes Molecular and biologic characterization and drug sensitivity regulated by structurally different histone deacetylase inhibitors. of pan-histone deacetylase inhibitor-resistant acute myeloid Proc Natl Acad Sci USA 2005; 102: 3697–3702. leukemia cells. Blood 2008; 112: 2896–2905. 28 Gui CY, Ngo L, Xu WS, Richon VM, Marks PA. Histone 45 Munshi A, Tanaka T, Hobbs ML, Tucker SL, Richon VM, deacetylase (HDAC) inhibitor activation of p21WAF1 involves Meyn RE. Vorinostat, a histone deacetylase inhibitor, enhances changes in promoter-associated proteins, including HDAC1. Proc the response of human tumor cells to ionizing radiation through Natl Acad Sci USA 2004; 101: 1241–1246. prolongation of gamma-H2AX foci. Mol Cancer Ther 2006; 5: 29 Archer SY, Meng S, Shei A, Hodin RA. p21(WAF1) is required for 1967–1974. butyrate-mediated growth inhibition of human colon cancer cells. 46 Ozaki K, Kishikawa F, Tanaka M, Sakamoto T, Tanimura S, Proc Natl Acad Sci USA 1998; 95: 6791–6796. Kohno M. Histone deacetylase inhibitors enhance the chemosen- 30 Huang L, Sowa Y, Sakai T, Pardee AB. Activation of the p21WAF1/ sitivity of tumor cells with cross-resistance to a wide range of CIP1 promoter independent of p53 by the histone deacetylase DNA-damaging drugs. Cancer Sci 2008; 99: 376–384. inhibitor suberoylanilide hydroxamic acid (SAHA) through the Sp1 47 Zhang Y, Adachi M, Zou H, Hareyama M, Imai K, Shinomura Y. sites. Oncogene 2000; 19: 5712–5719. Histone deacetylase inhibitors enhance phosphorylation of histone 31 Rosato RR, Almenara JA, Grant S. The histone deacetylase H2AX after ionizing radiation. Int J Radiat Oncol Biol Phys 2006; inhibitor MS-275 promotes differentiation or apoptosis in human 65: 859–866. leukemia cells through a process regulated by generation 48 Gaymes TJ, Padua RA, Pla M, Orr S, Omidvar N, Chomienne C of reactive oxygen species and induction of p21CIP1/WAF11. et al. Histone deacetylase inhibitors (HDI) cause DNA damage in Cancer Res 2003; 63: 3637–3645. leukemia cells: a mechanism for leukemia-specific HDI-dependent 32 Nebbioso A, Clarke N, Voltz E, Germain E, Ambrosino C, apoptosis? Mol Cancer Res 2006; 4: 563–573. Bontempo P et al. Tumor-selective action of HDAC inhibitors 49 Sanchez-Gonzalez B, Yang H, Bueso-Ramos C, Hoshino K, involves TRAIL induction in acute myeloid leukemia cells. Nat Quintas-Cardama A, Richon VM et al. Antileukemia activity of Med 2005; 11: 77–84. the combination of an anthracycline with a histone deacetylase 33 Davis T, Kennedy C, Chiew YE, Clarke CL, deFazio A. Histone inhibitor. Blood 2006; 108: 1174–1182. deacetylase inhibitors decrease proliferation and modulate cell 50 Rosato RR, Almenara JA, Maggio SC, Coe S, Atadja P, Dent P et al. cycle gene expression in normal mammary epithelial cells. Role of histone deacetylase inhibitor-induced reactive oxygen Clin Cancer Res 2000; 6: 4334–4342. species and DNA damage in LAQ-824/fludarabine antileukemic 34 Robbins AR, Jablonski SA, Yen TJ, Yoda K, Robey R, Bates SE et al. interactions. Mol Cancer Ther 2008; 7: 3285–3297. Inhibitors of histone deacetylases alter kinetochore assembly by 51 Chen CS, Wang YC, Yang HC, Huang PH, Kulp SK, Yang CC et al. disrupting pericentromeric heterochromatin. Cell Cycle 2005; 4: Histone deacetylase inhibitors sensitize prostate cancer cells to 717–726. agents that produce DNA double-strand breaks by targeting Ku70 35 Cimini D, Mattiuzzo M, Torosantucci L, Degrassi F. Histone acetylation. Cancer Res 2007; 67: 5318–5327. hyperacetylation in mitosis prevents sister chromatid separation 52 Hu Y, Lu W, Chen G, Zhang H, Jia Y, Wei Y et al. Overcoming and produces chromosome segregation defects. Mol Biol Cell resistance to histone deacetylase inhibitors in human leukemia 2003; 14: 3821–3833. with the redox modulating compound {beta}-phenylethyl isothio- 36 Qiu L, Burgess A, Fairlie DP, Leonard H, Parsons PG, Gabrielli BG. cyanate. Blood 2010; 116: 2732–2741. Histone deacetylase inhibitors trigger a G2 checkpoint in normal 53 Ungerstedt JS, Sowa Y, Xu WS, Shao Y, Dokmanovic M, Perez G cells that is defective in tumor cells. Mol Biol Cell 2000; 11: et al. Role of thioredoxin in the response of normal and 2069–2083. transformed cells to histone deacetylase inhibitors. Proc Natl Acad 37 Insinga A, Monestiroli S, Ronzoni S, Gelmetti V, Marchesi F, Viale Sci USA 2005; 102: 673–678. A et al. Inhibitors of histone deacetylases induce tumor-selective 54 Garcia-Manero G, Yang H, Bueso-Ramos C, Ferrajoli A, Cortes J, apoptosis through activation of the death receptor pathway. Wierda WG et al. Phase 1 study of the histone deacetylase Nat Med 2005; 11: 71–76. inhibitor vorinostat (suberoylanilide hydroxamic acid [SAHA]) in 38 Zhang XD, Gillespie SK, Borrow JM, Hersey P. The histone patients with advanced leukemias and myelodysplastic syn- deacetylase inhibitor suberic bishydroxamate regulates the dromes. Blood 2008; 111: 1060–1066. expression of multiple apoptotic mediators and induces mitochon- 55 Shao Y, Gao Z, Marks PA, Jiang X. Apoptotic and autophagic cell dria-dependent apoptosis of melanoma cells. Mol Cancer Ther death induced by histone deacetylase inhibitors. Proc Natl Acad 2004; 3: 425–435. Sci USA 2004; 101: 18030–18035. 39 Zhao Y, Tan J, Zhuang L, Jiang X, Liu ET, Yu Q. Inhibitors of 56 Hotchkiss RS, Strasser A, McDunn JE, Swanson PE. Cell death. histone deacetylases target the Rb-E2F1 pathway for apoptosis N Engl J Med 2009; 361: 1570–1583.

Leukemia HDAC inhibitors for MDS and AML A Quinta´s-Cardama et al 234 57 Wei Y, Kadia T, Tong W, Zhang M, Jia Y, Yang H et al. The 75 Kelly WK, Richon VM, O’Connor O, Curley T, MacGregor-Curtelli B, combination of a histone deacetylase inhibitor with the Bcl-2 Tong W et al. Phase I clinical trial of histone deacetylase inhibitor: homology domain-3 mimetic GX15-070 has synergistic antileu- suberoylanilide hydroxamic acid administered intravenously. kemia activity by activating both apoptosis and autophagy. Clin Clin Cancer Res 2003; 9 (Part 1): 3578–3588. Cancer Res 2010; 16: 3923–3932. 76 Kelly WK, O’Connor OA, Krug LM, Chiao JH, Heaney M, Curley T 58 Bali P, Pranpat M, Bradner J, Balasis M, Fiskus W, Guo F et al. et al. Phase I study of an oral histone deacetylase inhibitor, Inhibition of histone deacetylase 6 acetylates and disrupts the suberoylanilide hydroxamic acid, in patients with advanced chaperone function of heat shock protein 90: a novel basis for cancer. J Clin Oncol 2005; 23: 3923–3931. antileukemia activity of histone deacetylase inhibitors. J Biol Chem 77 Schaefer EW, Loaiza-Bonilla A, Juckett M, DiPersio JF, Roy V, 2005; 280: 26729–26734. Slack J et al. A phase 2 study of vorinostat in acute myeloid 59 Aoyagi S, Archer TK. Modulating molecular chaperone Hsp90 leukemia. Haematologica 2009; 94: 1375–1382. functions through reversible acetylation. Trends Cell Biol 2005; 78 Silverman LR, Verma A, Odchimar-Reissig R, LeBlanc A, 15: 565–567. Najfeld V, Gabrilove J et al. A phase I trial of the epigenetic 60 Cameron EE, Bachman KE, Myohanen S, Herman JG, Baylin SB. modulators vorinostat, in combination with azacitidine (azaC) in Synergy of demethylation and histone deacetylase inhibition in the patients with the myelodysplastic syndrome (MDS) and acute re-expression of genes silenced in cancer. Nat Genet 1999; 21: myeloid leukemia (AML): A Study of the New York Cancer 103–107. Consortium. Blood 2008; 112, (abstract 3656). 61 Gottlicher M, Minucci S, Zhu P, Kramer OH, Schimpf A, Giavara S 79 Kirschbaum M, Gojo I, Goldberg SL, Kujawski L, Atallah E, et al. Valproic acid defines a novel class of HDAC inhibitors Marks P et al. Vorinostat in combination with decitabine for the inducing differentiation of transformed cells. Embo J 2001; 20: treatment of relapsed or newly diagnosed acute myelogenous 6969–6978. leukemia (AML) or myelodysplastic syndrome (MDS): a phase I, 62 Cimino G, Lo-Coco F, Fenu S, Travaglini L, Finolezzi E, Mancini M Dose-Escalation Study. Blood 2009; 114 (abstract 2089). et al. Sequential valproic acid/all-trans retinoic acid treatment 80 Ravandi F, Faderl S, Thomas D, Burger J, Koller C, Garcia-Manero G reprograms differentiation in refractory and high-risk acute et al. Phase I study of sSuberoylanilide hydroxamic acid (SAHA) and myeloid leukemia. Cancer Res 2006; 66: 8903–8911. ecitabine in patients with relapsed, refractory or poor prognosis 63 Trus MR, Yang L, Suarez Saiz F, Bordeleau L, Jurisica I, Minden leukemia. Blood 2007; 110 (abstract 897). MD. The histone deacetylase inhibitor valproic acid alters 81 Shiozawa K, Nakanishi T, Tan M, Fang HB, Wang WC, Edelman MJ sensitivity towards all trans retinoic acid in acute myeloblastic et al. Preclinical studies of vorinostat (suberoylanilide hydro- leukemia cells. Leukemia 2005; 19: 1161–1168. xamic acid) combined with cytosine arabinoside and etoposide 64 Kuendgen A, Strupp C, Aivado M, Bernhardt A, Hildebrandt B, for treatment of acute leukemias. Clin Cancer Res 2009; 15: Haas R et al. Treatment of myelodysplastic syndromes with 1698–1707. valproic acid alone or in combination with all-trans retinoic acid. 82 Garcia-Manero G, Tambaro FP, Bekele BN, Jabbour E, Ravandi F, Blood 2004; 104: 1266–1269. Yang H et al. Phase II study of vorinostat in combination with 65 Kuendgen A, Knipp S, Fox F, Strupp C, Hildebrandt B, idarubicin (Ida) and cytarabine (ara-C) as front line therapy in Steidl C et al. Results of a phase 2 study of valproic acid acute myelogenous leukemia (AML) or higher risk myelodysplastic alone or in combination with all-trans retinoic acid in 75 patients syndrome (MDS). Blood 2009; 114 (abstract 1055). with myelodysplastic syndrome and relapsed or refractory 83 Fournel M, Bonfils C, Hou Y, Yan PT, Trachy-Bourget MC, Kalita A acute myeloid leukemia. Ann Hematol 2005; 84 (Suppl 1): et al. MGCD0103, a novel isotype-selective histone deacetylase 61–66. inhibitor, has broad spectrum antitumor activity in vitro and in 66 Bug G, Ritter M, Wassmann B, Schoch C, Heinzel T, Schwarz K vivo. Mol Cancer Ther 2008; 7: 759–768. et al. Clinical trial of valproic acid and all-trans retinoic acid in 84 Garcia-Manero G, Assouline S, Cortes J, Estrov Z, Kantarjian H, patients with poor-risk acute myeloid leukemia. Cancer 2005; 104: Yang H et al. Phase 1 study of the oral isotype specific histone 2717–2725. deacetylase inhibitor MGCD0103 in leukemia. Blood 2008; 112: 67 Kuendgen A, Schmid M, Schlenk R, Knipp S, Hildebrandt B, 981–989. Steidl C et al. The histone deacetylase (HDAC) inhibitor valproic 85 Garcia-Manero G, Yang AS, Klimek V, Cortes J, Ravandi F, acid as monotherapy or in combination with all-trans retinoic Newsome WM et al. Phase I/II Study of MGCD0103, an oral acid in patients with acute myeloid leukemia. Cancer 2006; 106: isotype-selective histone deacetylase (HDAC) inhibitor, in combi- 112–119. nation with 5-azacitidine in higher-risk myelodysplastic syndrome 68 Raffoux E, Chaibi P, Dombret H, Degos L. Valproic acid and all- (MDS) and acute myelogenous leukemia (AML). Blood 2007; 110 trans retinoic acid for the treatment of elderly patients with acute (abstract 444). myeloid leukemia. Haematologica 2005; 90: 986–988. 86 Gojo I, Jiemjit A, Trepel JB, Sparreboom A, Figg WD, Rollins S 69 Pilatrino C, Cilloni D, Messa E, Morotti A, Giugliano E, Pautasso M et al. Phase 1 and pharmacologic study of MS-275, a histone et al. Increase in platelet count in older, poor-risk patients deacetylase inhibitor, in adults with refractory and relapsed acute with acute myeloid leukemia or myelodysplastic syndrome treated leukemias. Blood 2007; 109: 2781–2790. with valproic acid and all-trans retinoic acid. Cancer 2005; 104: 87 Cheson BD, Bennett JM, Kopecky KJ, Buchner T, Willman CL, 101–109. Estey EH et al. Revised recommendations of the International 70 Soriano AO, Yang H, Faderl S, Estrov Z, Giles F, Ravandi F et al. Working Group for Diagnosis, standardization of response Safety and clinical activity of the combination of 5-azacytidine, criteria, treatment outcomes, and reporting standards for thera- valproic acid, and all-trans retinoic acid in acute myeloid peutic trials in acute myeloid leukemia. J Clin Oncol 2003; 21: leukemia and myelodysplastic syndrome. Blood 2007; 110: 4642–4649. 2302–2308. 88 Gore SD, Jiemjit A, Silverman LB, Aucott T, Baylin S, Carraway H 71 Garcia-Manero G, Kantarjian HM, Sanchez-Gonzalez B, Yang H, et al. Combined methyltransferase/histone deacetylase inhibition Rosner G, Verstovsek S et al. Phase 1/2 study of the combination of with 5-Azacitidine and MS-275 in patients with MDS, CMMoL and 5-aza-20-deoxycytidine with valproic acid in patients with AML: clinical response, histone acetylation and DNA damage. leukemia. Blood 2006; 108: 3271–3279. Blood 2006; 108 (abstract 517). 72 Issa J-P, Castoro R, Ravandi-Kashani F, Faderl S, Huang X, Estey E 89 Fandy TE, Herman JG, Kerns P, Jiemjit A, Sugar EA, Choi SH et al. et al. Randomized phase II study of combined epigenetic therapy: Early epigenetic changes and DNA damage do not predict clinical decitabine vs decitabine and valproic acid in MDS and AML. response in an overlapping schedule of 5-azacytidine and Blood 2008; 112 (abstract 228). entinostat in patients with myeloid malignancies. Blood 2009; 73 Marks PA, Miller T, Richon VM. Histone deacetylases. Curr Opin 114: 2764–2773. Pharmacol 2003; 3: 344–351. 90 Ellis L, Pan Y, Smyth GK, George DJ, McCormack C, Williams- 74 Mann BS, Johnson JR, Cohen MH, Justice R, Pazdur R. FDA Truax R et al. Histone deacetylase inhibitor panobinostat induces approval summary: vorinostat for treatment of advanced clinical responses with associated alterations in gene expression primary cutaneous T-cell lymphoma. Oncologist 2007; 12: profiles in cutaneous T-cell lymphoma. Clin Cancer Res 2008; 14: 1247–1252. 4500–4510.

Leukemia HDAC inhibitors for MDS and AML A Quinta´s-Cardama et al 235 91 Giles F, Fischer T, Cortes J, Garcia-Manero G, Beck J, Ravandi F 95 Byrd JC, Marcucci G, Parthun MR, Xiao JJ, Klisovic RB, Moran M et al. A phase I study of intravenous LBH589, a novel cinnamic et al. A phase 1 and pharmacodynamic study of depsipeptide hydroxamic acid analogue histone deacetylase inhibitor, in (FK228) in chronic lymphocytic leukemia and acute myeloid patients with refractory hematologic malignancies. Clin Cancer leukemia. Blood 2005; 105: 959–967. Res 2006; 12: 4628–4635. 96 Plumb JA, Finn PW, Williams RJ, Bandara MJ, Romero MR, 92 Furumai R, Matsuyama A, Kobashi N, Lee KH, Nishiyama M, Watkins CJ et al. Pharmacodynamic response and inhibition Nakajima H et al. FK228 (depsipeptide) as a natural prodrug that of growth of human tumor xenografts by the novel histone inhibits class I histone deacetylases. Cancer Res 2002; 62: 4916–4921. deacetylase inhibitor PXD101. Mol Cancer Ther 2003; 2: 93 Itoh Y, Suzuki T, Miyata N. Isoform-selective histone deacetylase 721–728. inhibitors. Curr Pharm Des 2008; 14: 529–544. 97 Gimsing P, Wu F, Qian X, Jeffers M, Knudsen L, Sehested M 94 Whittaker SJ, Demierre MF, Kim EJ, Rook AH, Lerner A, Duvic M et al. Activity of the histone deacetylase (HDAC) inhibitor et al. Final results from a multicenter, international, pivotal study of PXD101 in preclinical studies and in a phase I study in romidepsin in refractory cutaneous T-cell lymphoma. J Clin Oncol patients with advanced haematological tumors. Blood 2005; 106 2010; 28: 4485–4491. (abstract 3337).

Leukemia