Dual HDAC and PI3K Inhibitor CUDC-907 Downregulates MYC ... · Small Molecule Therapeutics Dual HDAC and PI3K Inhibitor CUDC-907 Downregulates MYC and Suppresses Growth of MYC-dependent

Small Molecule Therapeutics

Dual HDAC and PI3K Inhibitor CUDC-907Downregulates MYC and Suppresses Growthof MYC-dependent CancersKaiming Sun, Ruzanna Atoyan, Mylissa A. Borek, Steven Dellarocca,Maria Elena S. Samson, Anna W. Ma, Guang-Xin Xu, Troy Patterson,David P. Tuck, Jaye L. Viner, Ali Fattaey, and Jing Wang

Abstract

Upregulation of MYC is a common driver event in humancancers, and some tumors depend on MYC to maintaintranscriptional programs that promote cell growth and prolif-eration. Preclinical studies have suggested that individuallytargeting upstream regulators of MYC, such as histone deace-tylases (HDAC) and phosphoinositide 3-kinases (PI3K), canreduce MYC protein levels and suppress the growth of MYC-driven cancers. Synergy between HDAC and PI3K inhibition ininducing cancer cell death has also been reported, but theinvolvement of MYC regulation is unclear. In this study, wedemonstrated that HDAC and PI3K inhibition synergisticallydownregulates MYC protein levels and induces apoptosis in"double-hit" (DH) diffuse large B-cell lymphoma (DLBCL)cells. Furthermore, CUDC-907, a small-molecule dual-acting

inhibitor of both class I and II HDACs and class I PI3Ks,effectively suppresses the growth and survival of MYC-alteredor MYC-dependent cancer cells, such as DH DLBCL and BRD–

NUT fusion-positive NUT midline carcinoma (NMC) cells, andMYC protein downregulation is an early event induced byCUDC-907 treatment. Consistently, the antitumor activity ofCUDC-907 against multiple MYC-driven cancer types was alsodemonstrated in animal models, including DLBCL and NMCxenograft models,Myc transgenic tumor syngeneic models, andMYC-amplified solid tumor patient-derived xenograft (PDX)models. Our findings suggest that dual function HDAC andPI3K inhibitor CUDC-907 is an effective agent targeting MYCand thus may be developed as potential therapy for MYC-dependent cancers. Mol Cancer Ther; 16(2); 285–99. �2016 AACR.

IntroductionMYC, one of the most frequently deregulated oncogenes in

human cancer, encodes a transcription factor that plays a centralrole in tumor initiation and maintenance by orchestrating broadtranscriptional changes that facilitate growth and proliferation(1). Tumors can employmultiple genetic and epigenetic mechan-isms to upregulate MYC and commonly become dependent onMYC for proliferation and survival. Activating genetic alterationsofMYC, including genomic rearrangements, copy number gains,or amplifications, are common driving events in certain cancers.For example, acquisition of MYC alterations and subsequentupregulation ofMYC andMYC target genes have been implicatedas key events in the transformation of indolent follicular lym-phoma (FL) to highly aggressive diffuse large B-cell lymphoma(DLBCL) and is also observed in de novo DLBCL (2, 3). Alterna-tively, other oncogenic events drive cancer by inducing epigeneticchanges that promote MYC overexpression such as in NUTmidline carcinoma (NMC), a poorly differentiated, highly aggres-

sive subtype of squamous cell carcinoma that primarily arises inmidline organs (4). Approximately 80% of NMCs harbor achromosomal translocation that fuses NUT to BRD3 or BRD4,which encode bromodomain and extraterminal domain (BET)proteins that regulate gene expression in part through the seques-tration of p300 histone acetyltransferase activity and other tran-scriptionalmachinery to specific genomic loci (5).MYC is amajortarget of BET proteins, and NMC cells harboring BRD4–NUTfusions are dependent on MYC for maintenance of an undiffer-entiated, proliferative state (6, 7). Amplificationoroverexpressionof the MYC oncogene is also associated with poor prognosis inother solid tumor types, such as medulloblastoma (8, 9), breastcancer (10), ovarian cancer (11), prostate cancer (12), and lungcancer (13).

AlthoughMYC is anoncatalytic transcriptional regulator thathasthus far proven undruggable, increasing evidence suggests thattargeting upstream MYC regulators may be an effective strategy tosuppressMYC. For example, HDAC inhibitors have been shown tosuppressMYC transcription, in part by upregulating the transcrip-tion of MYC suppressor genes, such as FOXO1, in some (14–17)but not all cancer types (18). HDACs also facilitate MYC's onco-genic function as a transcription factor, as they are often recruited tothe promoter regions of some MYC-targeted genes to facilitateMYC-mediated regulation of transcription. HDAC inhibitors thusdecreaseMYC gene expression and restore the expression of genescoordinately suppressed by MYC family members and HDACs inmultiple cancer cell types, including DLBCL and NMC (19–25).

Blockade of the PI3K pathway has also been shown to suppressMYC activity by inhibiting MYC gene transcription (26) and

Curis, Inc., Lexington, Massachusetts.

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

Corresponding Author: Jing Wang, Curis, Inc. 4 Maguire Rd, Lexington,MA 02421. Phone: 617-645-3162; Fax: 617-503-6501; E-mail:[email protected].

doi: 10.1158/1535-7163.MCT-16-0390

�2016 American Association for Cancer Research.

MolecularCancerTherapeutics

www.aacrjournals.org 285

on January 17, 2020. © 2017 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst December 15, 2016; DOI: 10.1158/1535-7163.MCT-16-0390

http://mct.aacrjournals.org/

decreasing MYC protein stability. Activation of PI3K signalingeither through mutational activation of PI3K or through loss ofthe tumor suppressor PTEN leads to activation of the effectorkinase AKT, which in turns inhibits GSK3b through phosphory-lation (27). Active GSK3b normally phosphorylates MYC at thethreonine 58 (Thr58) residue, which facilitates the ubiquitin–proteasome degradation of MYC protein. PI3K inhibition thusdecreases MYC protein stability by releasing GSK3b from AKT-mediated inactivation and therefore promotingMYCThr58 phos-phorylation and subsequent degradation (28–31). Indeed, phar-macologic inhibition of the PI3K pathway is able to induce celldeath in PTEN-deficient DLBCL cells as a result of MYC down-regulation (32). PI3K inhibition also potentiates MYC down-regulation and cell death in MYC-dependent NMC cells (33).

Synergistic antitumor effect between HDAC and PI3K inhibi-tion has been reported in DLBCL xenograft tumors (34) and inMYC-drivenmousemedulloblastoma cells in vitro (16).However,the impact of this combination approach on MYC regulation hasnot been demonstrated. Because HDAC and PI3K pathways areboth involved in MYC regulation, we hypothesized that simul-taneously targeting these two pathwaysmay be an effective way tosuppress MYC.

CUDC-907 is an orally bioavailable small-molecule dualHDAC and PI3K inhibitor targeting class I and II HDACs andthe PI3Ka, b, and d isoforms. Its broad antitumor activities inhematologic and solid tumors have been previously reported byus (35). In a recent phase I study to evaluate the safety, tolerability,andpreliminary activity ofCUDC-907 inpatientswith relapsedorrefractory lymphoma or multiple myeloma, five of 37 (14%)evaluable patients achieved an objective response (two completeresponses and three partial responses), all of whom had DLBCL,including 3 patients with transformed FL (36). On the basis of thepreclinical evidence that PI3K and HDAC inhibitors can potentlysuppresses MYC and the confirmed clinical responses to CUDC-907 in a cancer type with frequent MYC alteration, we hypoth-esized that MYC-dependent cancer models would be responsiveto simultaneous HDAC and PI3K inhibition with CUDC-907.

In this study, we demonstrated that the combination of HDACand PI3K inhibition synergistically decreases MYC protein levelsand induces apoptosis in DHDLBCL cells. Moreover, CUDC-907potently suppresses MYC at least partially through decreasingMYC gene transcription and reducing MYC protein stability. Wealso demonstrate here the antitumor activity of CUDC-907 in vitroin DLBCL and NMC cells, and in vivo in MYC-altered DLBCL andMYC-dependent NMC xenograft models, Myc transgenic tumorsyngeneic models, andMYC-amplified solid tumor PDX models.Our results provide further evidence that CUDC-907 may havebroad utility in cancers that are driven by MYC upregulation.

Materials and MethodsReagents

CUDC-907 was synthesized in-house as described previously(35). Panobinostat (LBH-589), pictilisib (GDC-0941), I-BET-762, JQ1, OTX015, pan caspase inhibitor Z-VAD-FMK and pro-teasome inhibitor MG-132 were purchased from Selleck Chemi-cals. For in vitro assays, pan-caspase inhibitor Z-VAD-FMK andproteasome inhibitor MG-132 were dissolved in dimethyl sulf-oxide (DMSO) to generate 50 mmol/L stock solutions. Othercompounds were dissolved in DMSO to generate 1,000� stocksolutions and stored at �80�C for single use only. For in vivo

studies, CUDC-907 was formulated in 30% Captisol (LigandPharmaceuticals, Inc).

Cell cultureNMC cell lines were in-licensed from Dr. Christopher French

(Harvard Medical School, Boston, MA). No authentication wasdone by the authors. Other cancer cell lines were purchased fromthe ATCC (Manassas, VA) and German Collection ofMicroorgan-isms and Cell Cultures within one year of the study. No authen-tication was done by the authors. NMC cells were maintained inRPMI1640 þ GlutaMAX (Gibco by Life Technologies) mediumsupplemented with 10% FBS (Gibco) and incubated at 37�C in ahumidified atmosphere of 5% CO2. Other cell lines were main-tained as recommended by the source. Growth media werechanged every 2 to 3 days and cells were maintained at a densityof 2 � 106 to 4 � 106 cells/mL. Cells at exponential growth stagewere used for all experiments described below.

Cancer cell proliferation and caspase induction assaysCells were seeded at densities of 5 � 103 per well for prolif-

eration assay or 2 � 104 per well for caspase assay in 96-well flat-bottomed plates with the recommended culture medium. Cellswere then incubated with indicated compounds at various con-centrations for indicated amountof time in recommended growthmedium supplemented with 10% (v/v) FBS. Cell viability wasassessed using the CellTiter-Glo Luminescent Cell Viability Assay(Promega). Caspase-3/7 induction was assessed using the Cas-pase-Glo 3/7 Assay (Promega). GraphPad Prism 5.0 (GraphPadSoftware) was used for curve fitting and IC50 calculation. Twoindependent experiments with duplicates were performed foreach experiment.

Quantitative PCR experimentsCells were seeded at densities of 3 � 105 per mL in 24-well

plates with the recommended culture medium and incubatedwith indicated compounds at various concentrations for indicat-ed amount of time. Cells were then pelleted by centrifugation at10,000� g at 4�C for 10minutes, and immediately lysed in bufferRLT Plus (Qiagen). RNA was purified from lysed cells usingRNeasy columns (Qiagen) following the manufacturer's instruc-tions. cDNA was generated with SuperScript III First-Strand Syn-thesis SuperMix for qRT-PCR (Invitrogen by Life Technologies)according to the manufacturer's instructions. Quantitative PCR(qPCR) was performed using TaqMan Real-Time PCRMaster Mixand TaqMan Gene Expression Assay forMYC (Hs00153408_m1)and PPIA (Hs04194521_s1) on an Applied Biosystems qPCRinstrument (Applied Biosystems by Life Technologies). Relativeexpression was calculated with the comparative Ct method usinghousekeeping gene PPIA as loading control.

Western blot analysis and immunocytochemistrySnap-frozen tumor samples were pulverized with a mortar and

pestle on dry ice and transferred to tubes containing cold lysisbuffer [Halt phosphatase inhibitors (Thermo Scientific), Com-plete mini protease cocktail tablets (Roche Diagnostics), and T-PER Tissue Protein Extraction Reagent (Thermo Scientific)]. Onecold stainless steel bead (BioSpec Products) was placed into eachtube and the samples were then lysed using a Qiagen TissueLyser(Qiagen) in cold blocks. The samples were then centrifuged twiceat 16,000� g at 4�C for 15minutes. The samplesweredilutedwith

Sun et al.

Mol Cancer Ther; 16(2) February 2017 Molecular Cancer Therapeutics286




1� PBS buffer and the protein concentrations were measuredusing the BCA Protein Assay Kit (Thermo Scientific). Equalamount of total protein from each samples were loaded per laneonto NuPage gels (Invitrogen).

Cells were seeded at densities of 3 � 105 per mL in 24-wellplates with the recommended culture medium and incubatedwith indicated compounds at various concentrations for anindicated amount of time. Cells were then collected and centri-fuged at 10,000 � g at 4�C for 10 minutes. Cell pellets werewashed with cold 1� PBS and lysed and sheared in 1� SDSsample buffer (Sigma-Aldrich). Equal amounts of lysates fromeach sample were loaded per lane onto NuPage gels (Invitrogen).Following electrophoresis, samples were transferred to anitrocellulose membrane (Invitrogen). Membrane blocking andincubation with antibodies was performed using standard pro-cedures. Immunoblots were probed with the antibody of interest:anti-phospho-AKT (Ser473), anti-AKT(pan), anti-phospho-GSK-3b (Ser9), anti-GSK-3b, anti-phospho-p44/42 MAPK (ERK1/2)(Thr202/Tyr204), anti-p44/42 MAPK (ERK1/2), anti-BCL2, anti-MCL1, anti-c-MYC, anti-phospho-c-MYC (Ser62), anti-cleavedcaspase-3 (Asp 175), anti-Tubulin (Cell Signaling Technology);anti-phospho-c-MYC (Thr58; Thermo Scientific); anti-acetyl-His-tone H3 (Millipore). Membranes were cut based on the size ofmarkers and probed separately. Tubulin was used as loadingcontrol on each membrane, and markers from the samemembranes were boxed together in figures. IRDye 680- or800CW-conjugated secondary antibodies (LI-COR Biosci-ences) were used for detection. The membranes were scannedand the signal intensities were quantified on a LI-COR OdysseyInfrared Imaging System. The intensities of each marker werenormalized by the intensity of tubulin from the sample mem-brane. Ratios between phosphorylated and total proteins werecalculated according to the formula below using MYC as anexample, where pMYC represents intensity of phosphor-MYC;TubulinpMYC represents the intensity of tubulin from the samemembrane as pMYC; MYC represents intensity of total MYC;and TubulinMYC represents the intensity of tubulin from thesame membrane as the total MYC.

Ratio of MYC phosphorylation ¼ (pMYC / TubulinpMYC)/(MYC/TubulinMYC)

For dot Western blot analyses, a vacuum-based dot blot appa-ratus was used (Whatman Minifold I 96-well slot-blot arraysystem, Sigma-Aldrich). The lysates were spotted through circulartemplates directly onto the membrane. Membrane blocking andincubation with antibodies were performed using standard pro-tocols as described above for the Western blot assay. Tubulin wasused as the loading control. MYCprotein levels were calculated bythe intensity of MYC normalized by the intensity of tubulin fromthe same dot on the same membrane.

Combination synergy calculationThe combination synergy between panobinostat and picti-

lisib in MYC protein regulation was determined using twomathematic methods, Chou–Talalay equation and Bliss inde-pendent model. The MYC protein levels were calculated by theintensity of MYC normalized by the intensity of tubulin fromthe same dot blot on the same membrane as described above.Serial dilutions of panobinostat or pictilisib alone, or theircombination, were tested in a matrix format to get differentconstant molar ratios: 303:1, 100:1, 33:1, 11:1, 3.7:1, 1.2:1, or

0.41:1 of pictilisib and panobinostat. The combination index(CI) of each combination was calculated by the Chou–Talalayequation as described elsewhere (37) using CompuSyn soft-ware (38) (ComboSyn, Inc.).

For MYC protein level decrease, the Bliss independence modelis definedby the equationE(d1,d2)¼E(d1)þE(d2) - E(d1)�E(d2),where E(d1,d2) is the additive effect of panobinostat andpictilisibas predicted by their individual effects E(d1) and E(d2). Forcaspase-3/7 induction, the Bliss independence model is definedby the equation E(d1,d2)¼E(d1)þE(d2).

Differential gene expression analysisCells were pelleted by centrifugation at 10,000 � g at 4�C for

10 minutes, and immediately lysed in buffer RLT Plus (Qia-gen). RNA was purified from lysed cells using RNeasy columns(Qiagen) following the manufacturer's instructions. Total RNAinput was used to construct RNASeq libraries using the TruSeqRNA Sample Preparation Kit (Illumina) following the manu-facturer's instructions. Sequencing was performed on a HiSeq2500 System (Illumina). The bioconductor package Limma wasused for analysis of RNA-seq data (39, 40). RNA-seq counts for10 conditions (controlþ 2 dosesþ 2 timepoints for 2 cell lines)were preprocessed using the voom function (41), and linearmodels fit using the lmFit function. Specifically, a full modelwas fit (expression � 0 þ Treatment.group þ cellline.group þdose.group þ timepoint) and the coefficients extracted for theTreatment.group effect.

Two MYC target gene sets from the MSigDB Hallmark gene set"HALLMARK_MYC_TARGETS_V1" and "HALLMARK_MYC_-TARGETS_V2" were used for gene-set enrichment analysis (42).Clustering of genes was performed using complete linkage withEuclidean distance as a similarity metric. Correlations were com-puted using Spearman rank correlation test. The RNA-seq data aredeposited at GEO website, accession number GSE89339.

Lentivirus-mediated overexpression of MYC in WSU DLCL2cells

Cells were plated in 6-well tissue culture plate (BectonDickinson and Company) at a density of 7 � 104 cells per welland incubated at 37�C with 5% CO2 for 20–24 hours prior tolentivirus infection. Cells were then infected with CMV-C-MYClentivirus and vector control (Cellomics Technology) in poly-brene-containing medium (6 mg/mL, Sigma-Aldrich) and incu-bated at 37�Cwith 5%CO2. After overnight incubation, the CMV-C-MYC lentivirus-containing medium and vector control–con-taining medium were removed and replaced by fresh completegrowth medium. Cells continued to be incubated at 37�C with5% CO2.

In vivo tumor models and efficacy studiesEm-Myc mice (B6.Cg-Tg(IghMyc)22Bri/J) were obtained from

The Jackson Laboratory. The Em-Myc transgenics were monitoredweekly to identify any mice with malignant disease. Mice wereevaluated for any visible or palpable lumps, a hunched posture,tachypnea, a swollen belly, or ruffled fur and sacrificed promptlyupon the appearance of any such symptoms. Lymphomas thatemerged in lymph nodes were carefully dissected from sacrificedmice, washed in PBS, cut into < 3-mm pieces, and implanted intoC57BL/6J female mice (The Jackson Laboratory).

Six- to 9-week-old female immunodeficient athymic (nudenu/nu CD-1), Fox Chase SCID Beige, BALB/c nude mice, or

CUDC-907 as Potential Therapy for MYC-dependent Cancers

www.aacrjournals.org Mol Cancer Ther; 16(2) February 2017 287




MYC

/Tub

Nor

mal

ized

to

DM

SO C

ontr

ol

A

C

D E

Combina�on index

Pic�lisib (nmol/L)


Pano

bino

stat

(nm

ol/L

)

Pano

bino

stat

(nm

ol/L

)1.77 0.51 0.94 1.27 0.66 3.31.01 0.51 0.62 0.38 0.40 100.88 0.78 0.69 0.80 0.38 3012.3 37 111 333 1,000

0.3-0.7 Synergism0.7-0.85 Moderate synergism0.85-0.9 Slight synergism0.9-1.10 Nearly addi�ve1.1-1.2 Slight antagonism1.2-1.45 Moderate antagonism1.45-3.3 Antagonism

Control 12.3 37 111 333 1,000Pic�lisib

Pic�lisibPanobinostat_10 nmol/LPic�lisib+Panobinostat_10 nmol/LPredicted addi�ve effect

Pic�lisibPanobinostat_14 nmol/LPic�lisib+Panobinostat_14 nmol/LPredicted addi�ve effect

(nmol/L)Control 12.3 37 111 333 1,000

Control 3.3 10 30


MYC Protein Caspase-3/7

Concentra�on (mmol/L) Concentra�on (mmol/L)

Fold

cas

pase

-3/7

indu

c�on

norm

aliz

e to

DM

SO c

ontr

ol

MYC

/Tub

ulin

Panobinostat (nmol/L)

MYC Tubulin

B

Figure 1.

HDAC and PI3K inhibition synergistically downregulates MYC and induces caspase-3/7 activation in MYC-altered DLBCL cells. A, Dot Western blot analysis of MYCprotein (left) and tubulin (right) inWSU DLCL2 cells after 24 hours of treatment with various concentrations of HDAC inhibitor panobinostat, PI3K inhibitor pictilisib,or their combination at different concentrations as indicated. MYC and tubulin signalswere captured from the samemembrane at different fluorescencewavelength.B, Intensities of MYC signals are normalized by tubulin quantified from the dot Western blot analysis shown in A, and normalized to DMSO-treated cells. C,Combination indexes (CI) were calculated from the data presented in B using CompuSyn software. (Continued on the following page.)

Sun et al.





immunocompetent C57BL/6 mice obtained from Charles RiverLaboratorieswere either subcutaneously injectedwith 2� 106 to 3�106 cells in amediumsuspensionof100 to200mL (WSUDLCL2and 10-15 NMCmodels) or inoculated with tumor fragments (2-to 3-mm in diameter) collected from stock mice growing primaryhuman cancer tissues (ES2263, LU0377, and CR2506 models),U2932 tumors, or genetically engineered Em-Myc tumors (F2-3and M2-5) into the right hind flank region. The treatment wasstartedwhen the average tumor size reached approximately 100 to200 mm3 (WSU DLCL2, U2932, 10-15, ES2263, LU0377 andCR2506models) or one day after inoculation (F2-3 andM2-5 Em-Myc models). Varying doses of CUDC-907 or vehicle (30% Cap-tisol) were administered orally as per Institutional Animal Careand Use Committee guidelines. Tumor size was measured twiceweekly and the volumewas expressed inmm3using the formula,V¼ 0.5a� b2, where a and bwere the long and short diameters of thetumor, respectively. The tumor size was then used for the calcu-lation of tumor growth inhibition (TGI) ¼ [1 � (T1�T0)/(C1�C0)] � 100, where C1¼ mean tumor volume of controlmice at time t; T1¼mean tumor volume of treatedmice at time t;C0 ¼ mean tumor volume of control mice at time 0; and T0 ¼mean tumor volume of treated mice at time 0.

Statistical analysisDifferences between values obtained in tumors treated with

different experimental conditions were determined using theStudent t test on GraphPad Prism 5.0 (GraphPad Software).P < 0.05 was considered statistically significant.

ResultsHDAC and PI3K inhibition synergistically decreases MYCprotein levels and induces apoptosis

Because HDAC and PI3K pathways both regulate MYC levels,we evaluated the potential synergy between HDAC inhibitorpanobinostat (LBH-589) and PI3K inhibitor pictilisib (GDC-0941) in reducing MYC protein levels. In WSU DLCL2 DHDLBCL cells, synergistic suppression of MYC protein levels wasobserved when panobinostat was combined with pictilisib atmultiple concentrations with the best combination index (CI)at 0.38, indicating synergism, when evaluated with CompuSynsoftware (Fig. 1A–C). Synergy was also confirmed using theBliss independence model, showing that the combined effect isgreater than that predicted by their individual potencies (Fig.1D). Synergy between panobinostat and pictilisib in caspase-3/7 induction was also observed in the same cell line using theBliss independence method (Fig. 1E). These results suggestedthat HDAC and PI3K pathway inhibition synergistically reducesMYC protein levels and induces apoptosis in MYC-drivenDLBCL cells, providing a rationale for testing CUDC-907 inMYC-altered cancers.

CUDC-907 suppresses MYC protein, inhibits cell growth, andinduces apoptosis in MYC-altered DLBCL cell lines

On the basis of the synergy observed between HDAC and PI3Kinhibition in MYC regulation, we next evaluated the antitumoractivity of the dual functionHDACandPI3K inhibitorCUDC-907(chemical structure shown in Fig. 2A) in WSU DLCL2 cells, andobserved that CUDC-907 inhibits cell growth and induces apo-ptosis with nanomolar potency (Fig. 2B). The antitumor activityof CUDC-907 is more potent than the HDAC inhibitor panobi-nostat and the PI3K inhibitor pictilisib in a panel of DLBCL celllines tested (Fig. 2C and D), including cell lines with MYCalternation such as translocation, amplification, or overexpres-sion, and cell lines that harbor bothMYC andBCL2 translocations(DH), or MYC overexpression and BCL2 amplification ("double-expressor"), which have a particularly poor prognosis (43). Theantitumor activity of CUDC-907 appeared to be independent ofDLBCL cell-of-origin subtypes as both activated B cell-like (ABC)and germinal center B cell (GCB) type DLBCL cells were sensitiveto CUDC-907. Consistent with its potent effect in inhibiting cellgrowth and inducing apoptosis, CUDC-907 is also more effectivein downregulating MYC proteins in WSU DLCL2 cells, whencompared with pictilisib and panobinostat (Fig. 2E).

To understand the mechanism underlying the antitumor activ-ity of CUDC-907 in DH and "double expressor" DLBCL cells, weevaluated the changes induced by CUDC-907 in protein markersrelated to theHDAC, PI3K, andMYCpathways, including BCL2, aknown HDAC target gene (44) that cooperates with MYC toinduce lymphomagenesis (45), and MCL1, which is transcrip-tionally regulated by MYC (46) and whose downregulation hasbeen shown to contribute to synergy between HDAC and PI3Kinhibitors in DLBCL cells (47). As shown in Fig. 2F, 6 hours ofCUDC-907 treatment at nanomolar concentrations led to a dose-dependent accumulation of H3 acetylation (Ac-H3) and reduc-tion of PI3K pathway markers, such as phosphorylated AKT(pAKT) and phosphor-GSK3b, in WSU DLCL2 cells, indicatingpotent on-target inhibition of HDAC and PI3K activities, respec-tively. HDAC or PI3K inhibition is also induced by panobinostator pictilisib, respectively (Supplementary Fig. S1A–S1B). Dose-dependent reduction of MYC protein levels was observed 6 hourspost-CUDC-907 treatment (Fig. 2F). At the same time point, thelevels of MYC phosphorylation at the Ser62 site, which stabilizesthe MYC protein, decrease together with total MYC protein levels(Fig. 2F and G). However, no decrease was observed at the Thr58phosphorylation site, which targets MYC for ubiquitin–protea-some degradation, even though the total MYC protein levelsdecreased dramatically (Fig. 2F and G). In comparison, the levelsof BCL2 andMCL1 decrease after 24 hours of treatment, but not 6hours post-treatment (Fig. 2F). The levels of BIM, a proapoptoticBH3-only protein that is a negative regulator of BCL2, did notdecrease after 6 or 24 hours of treatment (Fig. 2F). Similar resultswere also observed in the "double-expressor" cell line U2932

(Continued.) Criteria for synergy, additivity, and antagonism are shown. D, The Bliss independencemethod was used to determine combinatorial activity between afixed concentration of panobinostat (10 nmol/L) and increasing concentrations of pictilisib. The effect of pictilisib on MYC protein level relative to tubulin isshown in red. The dashed green line represents the predicted additive effect calculated according to the Bliss independence method, whereas the solidgreen line shows the observed changes in MYC protein levels induced by the combination of panobinostat and pictilisib. E, The Bliss independencemethodwas usedto determine combinatorial activity between a fixed concentration of panobinostat (14 nmol/L) and increasing concentrations of pictilisib. The effect of pictilisib oncaspase activation is shown in red. The dashed green line represents the predicted additive effect calculated according to the Bliss independence method,whereas the solid green line shows the observed caspase-3/7 activity induced by the combination of panobinostat and pictilisib. Caspase-3/7 activation wasdetermined by the Caspase-Glo 3/7 assay after treatment with pictilisib and/or panobinostat for 24 hours. Data are normalized to DMSO-treated cells. Error barsrepresent SD of duplicates.






Sun et al.





harboring BCL2 amplification and MYC overexpression (Supple-mentary Fig. S1C and S1D). These results suggest that MYCreduction is an early event induced by CUDC-907 inMYC-alteredcell lines. Interestingly, CUDC-907 shifted the balance betweenthe two phosphorylation sites on MYC, Thr58 which promotesMYC protein degradation and Ser62 which stabilizes MYCprotein.

CUDC-907 decreases MYC gene transcription, induces MYCprotein degradation, and suppresses MYC function

To better understand the MYC regulation mechanism ofCUDC-907, we next evaluated the kinetics of MYC Thr58 phos-phorylation and MYC protein degradation in WSU DLCL2 cellstreated with nanomolar concentrations of CUDC-907 for varyinglengths of time ranging from 15 minutes to 6 hours. The MYCprotein levels start to decrease 2 hours post-CUDC-907 treatment,and continue decreasing until the last time point tested in thisexperiment, 6 hours (Fig. 3A). The accumulation of MYC Thr58phosphorylation, reflected by the ratios of phosphor-MYC(Thr58) to total MYC, also started after 2 hours of CUDC-907treatment, and this ratio increased about 12 folds after 6 hours oftreatment with 1 mmol/L CUDC-907 (Fig. 3B). The reduction inMYC protein levels induced by 6 or 24 hours of CUDC-907treatment is not attenuated by caspase inhibitor Z-VAD-FMK (Fig.3C), which inhibits the caspase activation induced by CUDC-907in these cells (Fig. 3C; Supplementary Fig. S2A and B). In fact, nocaspase activity was induced by 2 hours of CUDC-907 treatmentin these cells (Supplementary Fig. S2A). These results indicate thataccumulation of MYC Thr58 phosphorylation and caspase-inde-pendent downregulation ofMYC protein are early events inducedby CUDC-907.

Because bothHDACs andPI3K are involved in the transcriptionofMYC, we next evaluated changes inMYCmRNA levels inducedby CUDC-907 treatment. As shown in Fig. 3D, CUDC-907potently and dose dependently decreased MYC mRNA levels atnanomolar concentrations after 6 hours of treatment. HDACinhibitor panobinostat and PI3K inhibitor pictilisib also decreaseMYC mRNA levels, but are less potent than CUDC-907.

To further dissect the role of CUDC-907 in MYC regulation,MYC gene was expressed under the control of CMV promotermedicated by lentivirus transduction in WSU DLCL2 cells tobypass regulations on the c-MYC promoter. MYC protein levelincreased 0.5-fold as a result of the CMV-driven MYC expressionin the engineered cells (Fig. 3E), and the CMV-driven MYCexpression partially rescued the caspase activation (Fig. 3F) andcell growth inhibition (Supplementary Fig. S2C) induced by

CUDC-907. These results suggest that even though decreasedMYC transcription plays an important role in the antitumoractivity of CUDC-907, other factors may also contribute to itseffect.Moreover, after 2hours of incubation, bothCUDC-907 andPI3K inhibitor pictilisib decreased MYC protein levels in the cellsexpressing MYC under the control of the CMV promoter and thecells transduced with the control vector, whereas HDAC inhibitorpanobinostat induced no change in MYC levels at this early timepoint (Fig. 3G), suggesting that this early reduction of MYCprotein levels induced by CUDC-907 is not a result of MYCmRNA downregulation.

Because CUDC-907 induces accumulation of MYC phosphor-ylation at the Thr58 site, which targets MYC proteins to protea-somal degradation, we investigated the potential involvement ofproteasome in this early reduction inMYC protein levels inducedby CUDC-907. As shown in Fig. 3H, pharmacologic inhibition ofthe proteasome function by proteasome inhibitor MG-132 par-tially blocks the early reduction inMYC protein levels induced byCUDC-907 treatment in WSU DLCL2 cells, suggesting that pro-teasomal degradation ofMYC proteins is involved in this process.Collectively, our results indicate that both the transcriptionalregulation ofMYC gene expression and the proteasome-mediatedMYC protein degradation contribute to the MYC suppressioninduced by CUDC-907.

RNA-seq analysis was conducted to evaluate the effect ofCUDC-907 on MYC-associated genes. When WSU DLCL2 andDOHH2 DH DLBCL cells were treated with CUDC-907 at nano-molar concentrations for 6or 12hours, a total of 948differentiallyregulated geneswere identified (Supplementary Fig. S3A).Wefirstlooked for enrichment of curated MYC target genes within theCUDC-907–regulated 948-gene set. Of the 200 genes in the"HALLMARK_MYC_TARGETS_V1" gene set and the 58 genes inthe "HALLMARK_MYC_TARGETS_V2" gene set in the MolecularSignatures Database (MSigDB), 16 and 8 genes, respectively, arerepresented in the CUDC-907–regulated 948-gene set (Supple-mentary Fig. S3B). This corresponds to a 12.5-fold (Padj ¼1.67e�05) and 7.2-fold (Padj ¼ 5.59e�06) enrichment of MYCtarget genes in the CUDC-907–regulated gene set, respectively.

Second, we investigated whether CUDC-907 is causing expres-sion changes of genes whose mRNA levels are associated withMYC protein levels in DLBCL patients. We computed the corre-lation coefficients (r) between gene expression levels and MYCprotein levels across all samples in the TCGA DLBCL dataset. Forgenes in each bin defined by their correlation with MYC proteinlevels in DLBCL patients, we calculated the percentage of geneswhosemRNA levels are upregulated (logFC >0) or downregulated

Figure 2.CUDC-907 suppresses MYC protein, inhibits cell growth, and induces apoptosis in MYC-altered DLBCL cell lines. A, Chemical structure of CUDC-907. B, Dose–response curves of representative cell viability (left) and caspase-3/7 induction (right) in WSU DLCL2 cells treated with CUDC-907, panobinostat, or pictilisib.Data are normalized to DMSO-treated cells. Error bars represent SD of duplicates. C and D, Growth inhibition IC50 values (C) and caspase-3/7 induction EC50 values(D) of CUDC-907, the HDAC inhibitor panobinostat, or the PI3K inhibitor pictilisib in a panel of DLBCL cell lines are indicated. Cell viability was assessedby the CellTiter-Glo assay after 48-hour incubation with the indicated compound. Caspase-3/7 activation was determined by the Caspase-Glo 3/7 assay after24-hour incubation with the indicated compound. Growth inhibition IC50 values and caspase-3/7 induction EC50 values for each compound were determined withGraphPad Prism 5. The MYC status and the cell-of-origin of each cell line are indicated in the grafts. E, Dose–response curves of MYC protein reductioninduced byCUDC-907, panobinostat, or pictilisib inWSUDLCL2 cells. Data are represented as intensity ratios ofMYC to tubulin quantified from the samedotWesternblot analysis and normalized to DMSO-treated cells. IC50 values shown in parentheseswere calculated using GraphPad Prism 5. F, Immunoblot analysis of acetylatedHistone H3 (Ac-H3), phospho-AKT (pAKT), AKT, phospho-GSK3b (pGSK3b), GSK3b, phospho-ERK (pERK), ERK, phospho-MYC (Thr58 and Ser62), MYC, BCL2,MCL1, and BIM was performed on lysates of WSU-DLCL2 DLBCL cells treated with various concentrations of CUDC-907 for 6 or 24 hours, as indicated.Tubulin was used as a loading control. Individual blots are boxed together. G, Relative accumulation of phosphor-MYC (Thr58) and phosphor-MYC (Ser62) asrepresented by ratios of phosphor-MYC (normalized by tubulin from the same blot) to total MYC (also normalized by tubulin from the same blot)qualified from immunoblot analysis shown in F.






A B

C

ED

G

I

CUDC-907

Con

trol

1,00

0

111

12.3

1.4

CUDC-907

Con

trol

1,00

0

111

12.3

1.4

15 m

in

30 m

in

1 h 2 h

3 h

5 h

6 h

4 h

(Thr58)pMYC

MYC

Tubulin

(Thr58)pMYC

MYC

Tubulin

Tubulin

Tubulin

(Thr58)pMYC

MYC

Tubulin

Tubulin

(Thr58)pMYC

MYC

Tubulin

Tubulin

(nmol/L)

H

Con

trol

1,00

0

100

10 1 0.1

MYC

Tubulin

MG-132 (25 µmol/L)

2 h

Con

trol

1,00

0

100

10 1 0.1

CUDC-907 (nmol/L)CUDC-907 (nmol/L)

MYC

Cleaved Cas3

Tubulin

Con

trol

Vecto

rcM

YC

1,00

0

100

10 1 0.1

Z-VAD-FMK (10 mmol/L)CUDC-907 (nmol/L)CUDC-907 (nmol/L)

Con

trol

1,00

0

100

10 1 0.1

MYC

Cleaved Cas3

Tubulin 24

h

6 h

MYC

Tubulin

Con

trol

1,00

0

100

10 1 0.1

1,00

0

100

10 1 0.1

CUDC-907 (nmol/L) Panobinostat (nmol/L) Pictilisib (nmol/L)

MYC

Tubulin

Con

trol

1,00

0

100

10 1 0.1

Vector

CMV MYC

2 h

F

15 min

Fold

cas

pase

-3/7

indu

ctio

nno

rmal

ized

with

DM

SO

con

trol

Concentrations (mmol/L)

Drug concentration (nmol/L)

Rel

ativ

e M

YC

gen

e ex

pres

sion

1,00

00 1 4 12 37 111

333 0 1 3 10 30 0 1 4 12 37 111

333

1,00

0

0

2

4

6

pM

YC

(Th

r58)

/MY

C

8

10

12

30 min

DMSO

CUDC-907 1,000 nmol/LCUDC-907 111 nmol/LCUDC-907 12.3 nmol/L

CUDC-907 1.4 nmol/L

1 h 2 h 3 h 4 h 5 h 6 h

0.0

0.2

0.4

0.6

0.8

1.0

1.2CUDC-907 Panobinostat Pictilisib

Ratio ofMYC/Tubulin 1.52

Tubulin

MYC

1

5

4

3

2

1

00.000001 0.0001 0.01 1 100

Vector

Caspase 3/7 (24 hours)

cMYC

Sun et al.





(logFC <0) after CUDC-907 treatment (Fig. 3I). Genes with apositive correlation to MYC protein levels (r > 0.3) are morefrequently (�70%) downregulated after CUDC-907 treatment,whereas genes with a negative correlation to MYC protein levels(r < �0.3) are more frequently (�65%) upregulated afterCUDC-907 treatment. This result indicates that themRNA expres-sion changes induced by CUDC-907 in the two DLBCL cell linesare negatively correlated with correlation coefficients betweenexpression levels and MYC protein level in the TCGA DLBCLdataset (r ¼ �0.23; P ¼ 5.20e�218; Supplementary Fig. S3C),suggesting that CUDC-907 changes the mRNA expression of asubset of MYC-associated genes in DLBCL.

Together, these results indicate that CUDC-907 is able todownregulateMYCby inhibitingMYC transcription and reducingMYC protein stability, and it may also be able to at least partiallyreverse MYC-dependent transcriptional regulation of a subset ofMYC-associated genes in DLBCL.

Antitumor activity of CUDC-907 in MYC-driven DLBCLxenograft models and Em-Myc transgenic tumor syngeneicmodels of B-cell lymphoma

To determine whether CUDC-907 has activity against MYC-driven B-cell lymphomas in vivo, we evaluated the effect of CUDC-907 in severe combined immunodeficient beige (SCID-beige)mice bearing established WSU DLCL2 or U2932 xenografttumors. When orally administered at 100 mg/kg once daily everyday or 5 days on and 2 days off, CUDC-907 led to significantgrowth suppression of the WSU DLCL2 DH GCB DLBCL xeno-grafts and achieved near stasis in the "double expressor" U2932ABC DLBCL xenograft tumors, with tumor growth inhibition(TGI) of 69% and 97%, respectively, at the end of the efficacystudy (Fig. 4A and B) without causing severe body weight loss ornoticeable toxicity (Supplementary Fig. S4A-B).

To provide further confirmation that CUDC-907 has antitumoractivity against MYC-dependent B-cell lymphomas, we treatedC57BL/6 mice bearing Em-Myc transgenic tumors of B-celllymphoma (48) with CUDC-907 and found that CUDC-907markedly reduced tumor growth with a TGI of 72% and 58% intwomodels derived from tumors that spontaneously arose in twoEm-Myc transgenic mice (Fig. 4C and D) without inducing sig-nificant body weight loss or obvious toxicity (Supplementary

Fig. S4C–S4D). In addition, pharmacodynamic analysis of WSUDLCL2 xenograft tumors treated with 5 doses of CUDC-907 at100 mg/kg showed a dramatic reduction of MYC protein levels(Fig. 4E). SimilarMYCdownregulationwas also observed in othermodels, such as the Daudi Burkitt lymphoma model harboringMYC translocation (Supplementary Fig. S5), in which we previ-ously reported efficacy of CUDC-907 (35). These results demon-strate the antitumor efficacy of CUDC-907 and provide evidenceof CUDC-907–induced MYC downregulation in B-cell lympho-mas in vivo.

Antitumor activity of CUDC-907 inMYC-amplified solid tumorPDX models

To determine the effect of CUDC-907 in other MYC-driventumor models, CUDC-907 was evaluated in three PDX modelsharboring MYC gene amplifications, including an esophagealcancer with 12 copies of the MYC gene (ES2263), a non–smallcell lung cancer with 7 copies of the MYC gene (LU0377), and acolorectal cancer with 6 copies of the MYC gene (CR2506).Suppression of tumor growth was observed in response to treat-ment with CUDC-907 at 100 mg/kg 5-days-on and 2-days-off for21days in all threemodels tested,with TGIof 71%,54%, and54%in ES2263, LU0377, and CR2506, respectively (Fig. 5A–C), with-out causing significant body weight loss (data not shown). Theseresults demonstrate the broad antitumor activity of CUDC-907 inMYC-altered solid tumors.

CUDC-907 inhibits NMC growth in vitro and in vivoTo explore the antitumor activity of CUDC-907 inMYC-depen-

dent solid tumorswithout genetic alterations directly affecting theMYC gene, we next evaluated the efficacy of CUDC-907 in NMCcells harboring BRD–NUT fusions. We found that CUDC-907 ismore potent in inhibiting the growth of NMC cells in vitro thanBETbromodomain inhibitors (e.g., JQ1, I-BET-762, andOTX015)that directly target the BRD-NUT fusion protein and block growthof NMC cells by inducing MYC downregulation (6, 49; Fig. 6A).Similar results were also observed in twoMYC-driven DLBCL celllines,WSUDLCL2 andU2932, when treatedwithCUDC-907 andBET inhibitors (data not shown). In addition, CUDC-907 alsomore potently reduces the levels of MYC proteins (Fig. 6B) thaneither panobinostat or pictilisib in MYC-driven NMC cell lines

Figure 3.CUDC-907 decreases MYC transcription, induces MYC protein degradation, and suppresses MYC function in WSU DLCL2 DH DLBCL cells. A, Immunoblot analysisof phosphor-MYC (Thr58), total MYC, and tubulin was performed on lysates from the WSU DLCL2 cells treated with various concentrations of CUDC-907 for15minutes to6 hours. Tubulinwas used as a loading control. Individual blots are boxed together.B,Relative induction ofMYCThr58 phosphorylationwas representedas intensity ratios of phophos-MYC (normalized by tubulin from the same blot) to total MYC (also normalized by tubulin from the same blot) quantifiedfrom the immunoblot analysis shown inA. Concentrations of CUDC-907 and durations of treatment are as indicated.C, Immunoblot analysis ofMYC, cleaved caspase3, and tubulinwas performed on lysates from theWSUDLCL2 cells treatedwith various concentrations of CUDC-907with (right) orwithout (left) caspase inhibitor Z-VAD-FMK for 6 or 24 hours. Tubulin was used as loading control. Individual blots are boxed together. Concentrations of each compound and duration oftreatment are as indicated. D, Relative MYC gene expression in WSU DLCL2 cells treated with various concentrations of CUDC-907, panobinostat, or pictilisibfor 6 hours was calculated by the comparative Ct method using PPIA as a housekeeping gene. Concentrations of each compound are as indicated. E, Immunoblotanalysis of MYC and tubulinwas performed on lysates from theWSUDLCL2 cells infected by lentiviruses expressingMYC protein under the control of CMV promoteror control vector. Relative MYC protein levels represented as intensity ratios of MYC to tubulin are as indicated below the blot. F, Dose–response curve ofcaspase induction in WSU DLCL2 cells infected by lentiviruses expressing MYC protein under the control of CMV promoter or control vector after 24 hoursof CUDC-907 treatment. G, Immunoblot analysis of MYC and tubulin was performed on lysates from the WSU DLCL2 cells infected by lentivirus expressingMYC protein under the control of CMV promoter or control vector, and treated with various concentrations of CUDC-907, panobinostat, or pictilisib for 2 hours.Individual blots are boxed together. Concentrations of each compound are as indicated. H. Immunoblot analysis of MYC and tubulin was performed onlysates from the WSU DLCL2 cells treated with various concentrations of CUDC-907 with (right) or without (left) proteasome inhibitor MG-132 for 2 hours.Individual blots are boxed together. Concentrations of each compound are as indicated. I,CUDC-907 induced changes in themRNA expression pattern of a subset ofgenes whose mRNA levels are associated with MYC protein levels in the TCGA DLBCL dataset. The y-axis represents the percentage of genes whose mRNAexpression increases (logFC > 0) or decreases (logFC < 0) upon CUDC-907 treatment (FC ¼ fold change). The x-axis represents the Spearman correlationbetween each gene's mRNA levels and the MYC protein levels across all samples in the TCGA DLBCL dataset.






00 3 6 9

Days12 15 18 21 0 3 6 9

Days12 15

0 3 6 9

Days12 150

0

500

1,000

1,500

2,000

2,500

3,000

0

500

1,000

1,500

2,000

2,500

3,000

3 6 9

Days

Vehicle 1 h 3 h 6 h

CUDC-907

12 h 24 h

MYC

Tubulin

12 15

18 21

200

400

600

800

1,000

1,200

Tum

or

volu

me

(mm

3 )

Tum

or

volu

me

(mm

3 )

Tum

or

volu

me

(mm

3 )

Tum

or

volu

me

(mm

3 )

1,400

1,600

0

200

400

600

800

1,000

1,200

1,400

1,600

1,800

2,000

2,200

A B

C

E

D

Vehicle

WSU DLCL2

Em-Myc B-Cell lymphoma (F2-3) Em-Myc B-Cell lymphoma (M2-5)

U2932

Vehicle

CUDC-907 100 mg/kg PO QD 7/wk

Vehicle

CUDC-907 100 mg/kg PO QD 5/wkVehicle PO QD 5/wkCUDC-907 100 mpk PO QD 5/wk


n = 6P = 0.002

n = 10P = 0.0003

n = 9P = 0.00004

n = 9P = 0.0008

Figure 4.

Antitumor activity of CUDC-907 in humanDLBCL cell line–derived xenograftmodels and the Em-Myc transgenic tumormouse syngeneicmodels of B-cell lymphoma.A and B,Growth curves of establishedWSU DLCL2 (A) and U2932 (B) DLBCL xenograft tumors engrafted in severe combined immunodeficient beige (SCID-beige)mice upon oral administration of CUDC-907. Each graph describes the tumor volume (y-axis) as a function of the time after the initiation of treatment (x-axis).Each datapoint represents the mean volume of 6 (A) or 9 (B) independent tumors. Error bars represent SEM. WSU DLCL2 xenografts were treated withvehicle control (30% captisol) or CUDC-907 once a day for 21 days at 100 mg per kg body weight (mg/kg) as indicated. U2932 xenografts were treated withvehicle control (30% captisol) or CUDC-907 daily in a 5-days-on and 2-days-off schedule for 20 days at 100 mg/kg body weight as indicated. P values are asindicated. C andD, Growth curves of xenograft tumors derived from two independent B-cell lymphoma tumors spontaneously occurring in Em-Myc transgenic miceand engrafted in C57BL/6 mice upon oral administration of CUDC-907. Each graph describes the tumor volume (y-axis) as a function of the time after theinitiation of treatment (x-axis). Each datapoint represents the mean of 10 (C) or 9 (D) independent tumors. Error bars represent SEM. All mice were treated withvehicle control (30% captisol) or CUDC-907 daily at a 5-days-on, 2-days-off schedule for 12 (C) or 13 (D) days at 100 mg/kg body weight as indicated. P valuesare as indicated. E, Immunoblot analysis of MYC and tubulin was performed on lysates fromWSU DLCL2 xenograft tumors treated with 5 oral daily administrationsof vehicle control (30% captisol) or CUDC-907 at 100 mg/kg body weight as indicated. Three independent tumors at 1, 6, 12, and 24 hours and one tumor at3 hours were analyzed at each timepoint after the fifth administration as indicated. Individual blots are boxed together. Tubulin was used as a loading control.� , P < 0.05; �� , P < 0.01; �� , P < 0.001.


Sun et al.




(Supplementary Fig. S6). These results suggest that CUDC-907 isable to inhibit the growth of MYC-dependent NMC cells andreduce MYC protein levels.

In vivo, significant tumor growth suppression (TGI ¼ 94%; Fig.6C) and improved survival (Fig. 6D) were observed when BRD4–NUT fusion-positive NMC xenograft tumors were treated with100mg/kg of CUDC-907 on a 5-days-on and 2-days-off schedulefor 24 days or 63 days, respectively, without inducing significantbody weight loss or obvious toxicity (Supplementary Fig. S7).

These results indicate that CUDC-907 is effective in MYC-depen-dent NMCs.

Taken together, our results indicate that CUDC-907 effectivelydownregulates MYC protein levels in MYC-altered and MYC-dependent cells and tumor models, which at least is partiallyattributable to the synergy between simultaneous HDAC andPI3K inhibition inMYC regulation. These results provide a strongrationale for the clinical development of CUDC-907 in MYC-driven cancer indications.

00

500

1,000

1,500

2,000

2,500

3 6 9

Days

12 15 18 21

0 3 6 9

Days

12 15 18 21

0 3 6 9

Days

12 15 18 21

Tum

or

volu

me

(mm

3 )Tu

mo

r vo

lum

e (m

m3 )

n = 8P = 0.019

n = 8P = 0.004

n = 4P = 0.006

Vehicle

ES2263A

C

B

CR2506

LU0377


Vehicle


Vehicle

CUDC-907 100/75 mg/kg PO QD 5/wk

0

200

400

600

800

1,000

1,200

1,400Tu

mo

r vo

lum

e (m

m3 )

0

200

400

600

800

1,000

1,200

1,400

Figure 5.

CUDC-907 inhibits tumor growth in MYC-amplified patient-derived xenograft (PDX) models. Growth curves of established PDX tumors derived from (A)esophageal carcinoma (ES2263), (B) non–small cell lung carcinoma (LU0377), and (C) colorectal cancer (CR2506) engrafted in immunodeficient BALB/c nudemiceupon oral administration of CUDC-907. Each graph describes the tumor volume (y-axis) as a function of the time after the initiation of treatment (x-axis).Each datapoint represents the mean tumor volume of eight (A and B) or four (C) independent tumors. Error bars represent SEM. The mice from all threePDX models were treated with vehicle control (30% captisol) or CUDC-907 daily with a 5-days-on, 2-days-off schedule for 21 days at 100 mg/kg body weight asindicated. P values are as indicated. � , P < 0.05; �� , P < 0.01; �� , P < 0.001.






00-14

3

1032

6

10-15

0.1

1

10

100

1,000

10,000CUDC-907I-BET762

JQ1OTX015

IC50

(nm

ol/L

)A B

C D

**

Cont

rol

1,00

0

100

10

1 0.1

CUDC-907 (nmol/L)

10-15 cells 00-143 cells 10326 cells

MYC

pAKT

AKT

Ac-H3

Tubulin

Cont

rol

1,00

0

100

10

1 0.1

CUDC-907 (nmol/L)

Cont

rol

1,00

0

100

10

1 0.1

CUDC-907 (nmol/L)

E

Tum

or

volu

me

(mm

3 )

0 00 10 20 30 40 50 60 70

20

40P

erce

nt

of

surv

ival

60

80

100

120

0 3 6 9 12Days Days

15 18 21 24

200

400

600

800

1,000

1,200

1,400

1,600

1,800Vehicle

10-15 BRD4-NUT NMC 10-15 BRD4-NUT NMC


VehicleCUDC-907 100 mg/kg PO QD 5/wk

n = 5P = 0.004

Figure 6.

Antitumor activity of CUDC-907 against MYC-dependent NUT midline carcinoma (NMC) cells in vitro and in vivo. A, Growth inhibition IC50 values of CUDC-907 andthree BET inhibitors (I-BET-762, JQ1, and OTX015) in three BRD–NUT fusion-positive NMC cell lines. Cell viability after 72-hour incubation was assessedby the CellTiter-Glo assay. Growth inhibition IC50 values for each compound were determined by GraphPad Prism 5. (Continued on the following page.)


Sun et al.




DiscussionMYC-driven tumors, such as DLBCL, NMC, and other MYC-

amplified solid tumors, represent a significant unmet medicalneeddue to their poor prognosis, but efforts to directly targetMYChave not been successful to date. Here we show for the first timethat simultaneous inhibition of HDACs and PI3K, upstreamregulators of MYC gene expression and protein stability (Fig.6E), synergistically downregulatesMYCprotein levels. This obser-vation provided the foundation for testing the dual HDAC andPI3K inhibitor CUDC-907 inMYC-driven tumormodels. Indeed,our results showed that CUDC-907 potently reduce MYC proteinlevels in MYC-dependent cancers regardless of whether MYC isgenetically altered via translocation or amplification as in DHDLBCL and some solid tumor types, or epigenetically upregu-lated, as is the case in "double-expressor" DLBCL and NMC. Thispharmacodynamic effect was also confirmed in tumor tissuesfrom animal models, indicating that a sufficient therapeuticwindow exists in vivo. The broad antitumor activities of CUDC-907 in multiple in vivo models representing several indicationssuggest that it may be an effective therapy across a wide range ofhematologic and solid tumor types that are known to be depen-dent on MYC. Because of the homology between MYC familyproteins and potential similarities in their regulation, it will beinteresting to determine in future studieswhether CUDC-907 alsodownregulates the MYC homologs MYCN andMYCL and wheth-er CUDC-907 has activity in tumors where alterations in thesegenes are driving events. Given the pleiotropic effects of HDACinhibition and the multiple downstream effectors of PI3K, how-ever, it is likely that the effects of CUDC-907 are not completelyattributable to MYC suppression and that CUDC-907 will haveactivity in tumors caused by other genetic or epigenetic altera-tions. Indeed, CUDC-907 did suppress tumor growth of Karpas-422 xenografts that do not express high levels of MYC, albeit notto the same extent as MYC-overexpressing models (Supplemen-tary Fig. S8).

Additional studies are also needed to further dissect out theprimary and secondary effects of CUDC-907 to fully understandits broad impact on oncogenic networks. For example, wedetected relative accumulation of MYC Thr58 phosphorylationand caspase-independent reduction of MYC protein as early as 2hours after CUDC-907 treatment in both wild-type WSU DLCL2cells and the ones express MYC under the control of CMVpromoter, suggesting that this early reduction of MYC proteinmight be a primary result of the drug's effect on the PI3K pathwayinvolving no transcriptional regulation. In comparison, levels ofBCL2 and MCL1 remain unchanged until 6 to 24 hours post-treatment (Fig. 2F) when apoptosis starts (Supplementary Fig.S2A), suggesting that these may be secondary or subsequentdownstream effects of CUDC-907 treatment. One may suspect

that the decrease of BCL2 and MCL1 levels might be part of themassive protein degradation that typically occurs during the latestage of apoptosis. However, it appears not to be the case forCUDC-907 in this setting, as the levels of the proapoptotic proteinBIMkept increasing throughout the 24-hour treatment, indicatingspecific drug effects instead of massive protein degradation.Thorough studies focusing on the mechanism of CUDC-907'santitumor actionsmayhelp to identify other sensitive tumor typesin addition to the MYC-driven cancers.

Regardless of the total contribution of MYC suppression to theantitumor activity ofCUDC-907, our preclinicalfindings showingthat CUDC-907 effectively suppresses MYC and has activity inmultiple models of MYC-driven cancers suggests that patientswhose tumors harbor genetic alterations inMYC or are dependenton MYC overexpression may benefit from CUDC-907 treatment.This is consistent with the results of the recently published phase Istudy of CUDC-907 in patients with relapsed or refractory lym-phoma or multiple myeloma. In the interim analysis of this trial,the single-agent activity of CUDC-907 was observed in patientswith DLBCL and transformed FL, which is commonly driven byMYC overexpression, amplification, or translocation, and led totwo complete and three partial responses (36). To our knowledge,single-target HDAC or PI3K inhibitors have not shown suchpromising efficacy in this population. Moreover, CUDC-907 isgenerally well tolerated, causing manageable gastrointestinal andhematologic events that were similar to those observed with FDA-approved single-target HDAC and PI3K inhibitors without theserious adverse events observed with HDAC or PI3K inhibitorssuch as infections, colitis, hepatotoxicity, and cardiotoxicity (36).Of note, the combination of the MTDs of the FDA-approvedHDAC inhibitors, vorinostat or panobinostat and the FDA-approved PI3K inhibitor idealisib are not tolerable in animalmodels in our experience (33 and data not shown), suggestingthat such combination strategies may not be feasible in patients.Although longer follow-up studies and additional randomizedstudies in larger cohorts are needed, the favorable safety profileand clinical efficacy of CUDC-907 suggests that this compoundwould have numerous advantages over combination treatmentwith single-targetHDACandPI3K inhibitors. Even though a dual-function single agent may lose the flexibility of independent doseadjusting, the clinical development process of this type of dual-function drug candidate is much simpler than combinationtherapy because the pharmacokinetics and toxicity profile aremore predictable and easier to manage.

Based in part on these preclinical findings, a phase I study toevaluate the safety, tolerability, and pharmacokinetics of CUDC-907 has been initiated in patients with advanced/relapsed solidtumors, including NMC (NCT02307240), and given the encour-aging clinical results in patients with DLBCL, a biomarker-driven

(Continued.) B, Immunoblot analysis of MYC, pAKT, AKT, and Ac-H3 was performed on lysates from three BRD–NUT fusion-positive cell lines after treatment withvarious concentrations of CUDC-907 for 24 hours as indicated. Tubulin was used as a loading control. Individual blots are boxed together. C, Tumor growthcurves of 10-15 NMC cell line–derived xenograft tumors engrafted in immunodeficient athymic nude mice upon oral administration of CUDC-907. The y-axisshows the tumor volume as a function of the time after the initiation of treatment (x-axis). Each data point represents the mean tumor volume of five independenttumors. Error bars represent SEM. Mice were treated with vehicle control (30% captisol) or CUDC-907 on a 5-days-on, 2-days-off schedule for 24 days at100 mg/kg body weight as indicated. The vehicle group had to be terminated earlier (day 14) because the tumor sizes exceeded institutional limitations. D, Animalsurvival curves of 10-15 NMC cell line–derived xenograft tumors engrafted in immunodeficient athymic nudemice upon oral administration of CUDC-907. The y-axisshows the percentage of live animals as a function of time after initiation of treatment (x-axis). Micewere treatedwith vehicle control (30% captisol) or CUDC-907 ona 5-days-on, 2-days-off schedule for 63 days at 100 mg/kg body weight as indicated. E, Proposed mechanism of MYC suppression induced by CUDC-907.Simultaneous HDAC and PI3K inhibition by CUDC-907 suppresses MYC through transcriptional and post-translational mechanisms and thus leads to greaterinhibition of MYC than single-target HDAC or PI3K inhibitors.






phase II trial (NCT02674750) has also been initiated to evaluatethe activity of CUDC-907 with or without rituximab in patientswithMYC-alteredDLBCL inwhich the enrollment criteria includerelapsed or refractory DLBCL patients who have MYC transloca-tion, amplification, overexpression, or copy number gain. Futurestudies will explore other genetic and epigenetic backgrounds inwhich CUDC-907 might be effective as well as potential ways toutilize CUDC-907 in combination with other antitumor agents.

Disclosure of Potential Conflicts of InterestA. Fattaey has ownership interest (including patents) in Curis, Inc. J. Wang is

a director at TesaroBio. No potential conflicts of interest were disclosed by theother authors.

Authors' ContributionsConception and design: K. Sun, A. Fattaey, J. WangDevelopment of methodology: K. Sun, T. Patterson, J. Wang

Acquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): R. Atoyan, M.A. Borek, S. DellaRocca, M.E.S. Samson,G.-X. Xu, T. Patterson, J.L. Viner, J. WangAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): K. Sun, R. Atoyan, M.A. Borek, S. DellaRocca, G.-X.Xu, T. Patterson, D.P. Tuck, J. WangWriting, review, and/or revisionof themanuscript:K. Sun,A.W.Ma,D.P. Tuck,J.L. Viner, A. Fattaey, J. WangAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): S. DellaRocca, M.E.S. SamsonStudy supervision: K. Sun, J. Wang

Grant SupportThis study was funded by Curis Inc.The costs of publication of this articlewere defrayed inpart by the payment of

page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received June 16, 2016; revised October 26, 2016; accepted November 12,2016; published OnlineFirst December 15, 2016.

References1. Gabay M, Li Y, Felsher DW. MYC activation is a hallmark of cancer

initiation and maintenance. Cold Spring Harb Perspect Med 2014;4.2. Ott G, Rosenwald A, Campo E. Understanding MYC-driven aggressive B-cell

lymphomas: pathogenesis and classification. Blood 2013;122:3884–91.3. Pasqualucci L, Khiabanian H, Fangazio M, Vasishtha M, Messina M,

Holmes AB, et al. Genetics of follicular lymphoma transformation. CellRep 2014;6:130–40.

4. French CA.The importance of diagnosing NUT midline carcinoma. HeadNeck Pathol 2013;7:11–6.

5. Shi J, Vakoc CR. The mechanisms behind the therapeutic activity of BETbromodomain inhibition. Mol Cell 2014;54:728–36.

6. Delmore JE, Issa GC, Lemieux ME, Rahl PB, Shi J, Jacobs HM, et al. BETbromodomain inhibition as a therapeutic strategy to target c-Myc. Cell2011;146:904–17.

7. Grayson AR, Walsh EM, Cameron MJ, Godec J, Ashworth T, Ambrose JM,et al.MYC, a downstream target of BRD-NUT, is necessary and sufficient forthe blockade of differentiation in NUT midline carcinoma. Oncogene2014;33:1736–42.

8. Cho YJ, Tsherniak A, Tamayo P, Santagata S, Ligon A, Greulich H, et al.Integrative genomic analysis of medulloblastoma identifies a molecularsubgroup that drives poor clinical outcome. J Clin Oncol 2011;29:1424–30.

9. Northcott PA, Korshunov A,Witt H,Hielscher T, Eberhart CG,Mack S, et al.Medulloblastoma comprises four distinct molecular variants. J Clin Oncol2011;29:1408–14.

10. Xu J, Chen Y, Olopade OI. MYC and breast cancer. Genes Cancer2010;1:629–40.

11. Dimova I, Raitcheva S, Dimitrov R, Doganov N, Toncheva D. Correlationsbetween c-myc gene copy-number and clinicopathological parameters ofovarian tumours. Eur J Cancer 2006;42:674–9.

12. Koh CM, Bieberich CJ, Dang CV, Nelson WG, Yegnasubramanian S, DeMarzo AM. MYC and prostate cancer. Genes Cancer 2010;1:617–28.

13. Seo AN, Yang JM, Kim H, Jheon S, Kim K, Lee CT, et al. Clinicopathologicand prognostic significance of c-MYC copy number gain in lung adeno-carcinomas. Br J Cancer 2014;110:2688–99.

14. Bouchard C, Marquardt J, Bras A, Medema RH, Eilers M. Myc-inducedproliferation and transformation require Akt-mediated phosphorylationof FoxO proteins. EMBO J 2004;23:2830–40.

15. Peart MJ, Smyth GK, van Laar RK, Bowtell DD, Richon VM,Marks PA, et al.Identification and functional significance of genes regulated by structurallydifferent histone deacetylase inhibitors. Proc Natl Acad Sci U S A2005;102:3697–702.

16. Pei Y, Liu KW, Wang J, Garancher A, Tao R, Esparza LA, et al. HDAC andPI3K antagonists cooperate to inhibit growth of MYC-drivenmedulloblas-toma. Cancer Cell 2016;29:311–23.

17. Zain J, O'Connor OA. Targeting histone deacetyalses in the treatment of B-and T-cell malignancies. Invest New Drugs 2010;28:S58–78.

18. Bhadury J, Nilsson LM, Muralidharan SV, Green LC, Li Z, Gesner EM, et al.BET and HDAC inhibitors induce similar genes and biological effects andsynergize to kill in Myc-induced murine lymphoma. Proc Natl Acad SciU S A 2014;111:E2721–30.

19. Kurland JF, TanseyWP.Myc-mediated transcriptional repressionby recruit-ment of histone deacetylase. Cancer Res 2008;68:3624–9.

20. Zhang X, Zhao X, FiskusW, Lin J, Lwin T, RaoR, et al. Coordinated silencingof MYC-mediated miR-29 by HDAC3 and EZH2 as a therapeutic target ofhistone modification in aggressive B-Cell lymphomas. Cancer Cell2012;22:506–23.

21. Polack A, Eick D, Koch E, Bornkamm GW. Truncation does not abrogatetranscriptional downregulation of the c-myc gene by sodium butyrate inBurkitt's lymphoma cells. EMBO J 1987;6:2959–64.

22. Chambers AE, Banerjee S, Chaplin T, Dunne J, Debernardi S, Joel SP, et al.Histone acetylation-mediated regulation of genes in leukaemic cells. Eur JCancer 2003;39:1165–75.

23. Gui CY, Ngo L, Xu WS, Richon VM, Marks PA. Histone deacetylase(HDAC) inhibitor activation of p21WAF1 involves changes in promot-er-associated proteins, including HDAC1. Proc Natl Acad Sci U S A2004;101:1241–6.

24. Liu T, Tee AE, Porro A, Smith SA, Dwarte T, Liu PY, et al. Activation of tissuetransglutaminase transcription by histone deacetylase inhibition as atherapeutic approach for Myc oncogenesis. Proc Natl Acad Sci U S A2007;104:18682–7.

25. Van Lint C, Emiliani S, Verdin E. The expression of a small fraction ofcellular genes is changed in response to histone hyperacetylation. GeneExpr 1996;5:245–53.

26. Dey N, Leyland-Jones B, De P. MYC-xing it up with PIK3CA mutation andresistance to PI3K inhibitors: summit of two giants in breast cancers. Am JCancer Res 2015;5:1–19.

27. Cross DA, Alessi DR, Cohen P, Andjelkovich M, Hemmings BA. Inhibitionof glycogen synthase kinase-3 by insulin mediated by protein kinase B.Nature 1995;378:785–9.

28. Kenney AM, Widlund HR, Rowitch DH. Hedgehog and PI-3 kinase sig-naling converge on Nmyc1 to promote cell cycle progression in cerebellarneuronal precursors. Development 2004;131:217–28.

29. Asano T, Yao Y, Zhu J, Li D, Abbruzzese JL, Reddy SA. The PI 3-kinase/Aktsignaling pathway is activated due to aberrant Pten expression and targetstranscription factors NF-kappaB and c-Myc in pancreatic cancer cells.Oncogene 2004;23:8571–80.

30. Kumar A, Marques M, Carrera AC. Phosphoinositide 3-kinase activation inlate G1 is required for c-Myc stabilization and S phase entry. Mol Cell Biol2006;26:9116–25.


Sun et al.




31. Bechard M, Dalton S. Subcellular localization of glycogen synthase kinase3beta controls embryonic stem cell self-renewal. Mol Cell Biol 2009;29:2092–104.

32. PfeiferM, GrauM, Lenze D,Wenzel SS,Wolf A,Wollert-Wulf B, et al. PTENloss defines a PI3K/AKT pathway-dependent germinal center subtype ofdiffuse large B-cell lymphoma. Proc Natl Acad Sci U S A 2013;110:12420–5.

33. Tinsley S, Meja K, Shepherd C, Khwaja A. Synergistic induction of celldeath in haematological malignancies by combined phosphoinositide-3-kinase and BET bromodomain inhibition. Br J Haematol 2015;170:275–8.

34. Gupta M, Ansell SM, Novak AJ, Kumar S, Kaufmann SH, Witzig TE.Inhibition of histone deacetylase overcomes rapamycin-mediated resis-tance in diffuse large B-cell lymphoma by inhibiting Akt signaling throughmTORC2. Blood 2009;114:2926–35.

35. Qian C, Lai CJ, Bao R, Wang DG, Wang J, Xu GX, et al. Cancer networkdisruption by a single molecule inhibitor targeting both histone deacety-lase activity and phosphatidylinositol 3-kinase signaling. Clin Cancer Res2012;18:4104–13.

36. Younes A, Berdeja JG, Patel MR, Flinn I, Gerecitano JF, Neelapu SS, et al.Safety, tolerability, and preliminary activity of CUDC-907, a first-in-class,oral, dual inhibitor of HDAC and PI3K, in patients with relapsed orrefractory lymphoma or multiple myeloma: an open-label, dose-escala-tion, phase 1 trial. Lancet Oncol 2016;17:622–31.

37. Chou TC, Talalay P. Quantitative analysis of dose-effect relationships: thecombined effects of multiple drugs or enzyme inhibitors. Adv EnzymeRegul 1984;22:27–55.

38. Chou TC, Martin N. CompuSyn for drug combinations and for generaldose-effect analysis; 2005.

39. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powersdifferential expression analyses for RNA-sequencing and microarraystudies. Nucleic Acids Res 2015;43:e47.

40. Smyth GK.Linear models and empirical bayes methods for assessingdifferential expression in microarray experiments. Stat Appl Genet MolBiol 2004;3:Article3.

41. Law CW, Chen Y, Shi W, Smyth GK. voom: Precision weights unlock linearmodel analysis tools for RNA-seq read counts. Genome Biol 2014;15:R29.

42. SubramanianA, TamayoP,Mootha VK,Mukherjee S, Ebert BL,GilletteMA,et al. Gene set enrichment analysis: a knowledge-based approach forinterpreting genome-wide expression profiles. Proc Natl Acad Sci U S A2005;102:15545–50.

43. Sarkozy C, Traverse-Glehen A, Coiffier B. Double-hit and double-protein-expression lymphomas: aggressive and refractory lymphomas. LancetOncol 2015;16:e555–67.

44. Duan H, Heckman CA, Boxer LM. Histone deacetylase inhibitors down-regulate bcl-2 expression and induce apoptosis in t(14;18) lymphomas.Mol Cell Biol 2005;25:1608–19.

45. Strasser A,Harris AW, BathML, Cory S. Novel primitive lymphoid tumoursinduced in transgenic mice by cooperation betweenmyc and bcl-2. Nature1990;348:331–3.

46. Labisso WL, Wirth M, Stojanovic N, Stauber RH, Schnieke A, Schmid RM,et al. MYC directs transcription of MCL1 and eIF4E genes to controlsensitivity of gastric cancer cells toward HDAC inhibitors. Cell Cycle2012;11:1593–602.

47. Rahmani M, Aust MM, Benson EC, Wallace L, Friedberg J, Grant S. PI3K/mTOR inhibition markedly potentiates HDAC inhibitor activity in NHLcells throughBIM- andMCL-1-dependentmechanisms in vitro and in vivo.Clin Cancer Res 2014;20:4849–60.

48. Adams JM, Harris AW, Pinkert CA, Corcoran LM, Alexander WS, Cory S,et al. The c-myc oncogene driven by immunoglobulin enhancers induceslymphoid malignancy in transgenic mice. Nature 1985;318:533–8.

49. Coude MM, Braun T, Berrou J, Dupont M, Bertrand S, Masse A, et al. BETinhibitor OTX015 targets BRD2 and BRD4 and decreases c-MYC in acuteleukemia cells. Oncotarget 2015;6:17698–712.






2017;16:285-299. Published OnlineFirst December 15, 2016.Mol Cancer Ther Kaiming Sun, Ruzanna Atoyan, Mylissa A. Borek, et al. Suppresses Growth of MYC-dependent CancersDual HDAC and PI3K Inhibitor CUDC-907 Downregulates MYC and

Updated version

10.1158/1535-7163.MCT-16-0390doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2016/12/13/1535-7163.MCT-16-0390.DC1

Access the most recent supplemental material at:

Cited articles

http://mct.aacrjournals.org/content/16/2/285.full#ref-list-1

This article cites 47 articles, 18 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/16/2/285.full#related-urls

This article has been cited by 4 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mct.aacrjournals.org/content/16/2/285To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-16-0390

http://mct.aacrjournals.org/content/suppl/2016/12/13/1535-7163.MCT-16-0390.DC1

http://mct.aacrjournals.org/content/16/2/285.full#ref-list-1

http://mct.aacrjournals.org/content/16/2/285.full#related-urls

http://mct.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mct.aacrjournals.org/content/16/2/285


Dual HDAC and PI3K Inhibitor CUDC-907 Downregulates MYC ... · Small Molecule Therapeutics Dual HDAC and PI3K Inhibitor CUDC-907 Downregulates MYC and Suppresses Growth of MYC-dependent

Documents