+ All Categories
Home > Documents > EZH2 inhibition as an effective treatment for metastatic ... · EZH2 inhibition is effective in...

EZH2 inhibition as an effective treatment for metastatic ... · EZH2 inhibition is effective in...

Date post: 17-Jul-2020
Category:
Upload: others
View: 1 times
Download: 0 times
Share this document with a friend
1
0.1 1 10 LNCaP 22Rv1 VCaP PC3 DU145 GI50 ( µ M CPI-1205) 0% 50% 100% 150% 200% D4 D7 D11D14 Cell viability Gene sets enriched : GO_TISSUE_DEVELOPMENT GO_EPITHELIUM_DEVELOPMENT GO_REGULATION_OF_CELL_DIFFERENTIATION HALLMARK_ANDROGEN_RESPONSE NELSON_RESPONSE_TO_ANDROGEN_UP BENPORATH_PROLIFERATION EZH2 inhibition as an effective treatment for metastatic castration-resistant prostate cancer William D. Bradley¹, John P. McGrath¹, CC Yuan¹, Feng Zhao¹, Priyanka Sawant¹, Charlie Hatton², Xinwei Han², Barbara Bryant², Andrew Conery², and Patrick Trojer¹ Translational Sciences¹, Functional Genomics², Constellation Pharmaceuticals, 215 First Street, Cambridge, MA, USA 02142 Introduction 0 50 100 150 200 0.01 0.1 1 10 LNCaP-AR enzalutamide (μM) Relative cell viability (%) - doxycycline + doxycycline AR ACTB LNCaP-EV LNCaP-AR - 500 - 10 50 100 500 1000 ng/ml doxycycline Day Dox 22Rv1 EED KO 0 + + + - - - 4 8 14 15 18 0 + - 8 22Rv1 EED WT - H L LNCaP +CPI-1205 - H L LNCaP +CPI-1205 0 2 row z-score -2 NELSON_RESPONSE_TO_ANDROGEN_UP HALLMARK_ANDROGEN_RESPONSE WONG_ADULT_TISSUE_STEM_MODULE SCHAEFFER_PROSTATE_DEVELOPMENT_48HR_UP NUYTTEN_EZH2_TARGETS_UP HALLMARK_E2F_TARGETS GO_REGULATION_OF_CELL_PROLIFERATION NUYTTEN_EZH2_TARGETS_DN Gene sets enriched: enzalutamide CPI-1205 + - + + LNCaP - - + - Enhancement of enzalutamide effect Maintenance of enzalutamide effect Maintenance of CPI-1205 effect EZH2 is the catalytic subunit of the PRC2 complex, and methylates histone H3 on lysine 27, resulting in suppression of gene transcription EZH2 mutations and increased expression are often observed in cancer, leading to repression of genes associated with apoptosis and differentiation (1-2) Over-expression in metastatic castration-resistant prostate cancer (mCRPC [3]), and decreased expression of target genes correlate with poor prognosis (4) While accounting for a small percentage of the overall disease incidence (5), mCRPC is very aggressive with high morbidity (6) Many mCRPC tumors remain dependent on androgen receptor signaling (ARS), but eventually develop resistance to ARS inhibitors, such as enzalutamide and abiraterone acetate Recent studies have linked EZH2 inhibition to re-sensitization to ARS inhibitors in neuroendocrine prostate cancer (NEPC) (7-9), and other PC models (10) Given this unmet need, and the link between EZH2 and advanced prostate cancer, we sought to determine if EZH2 inhibition can be utilized as an effective treatment and/or be combined with ARS inhibitor therapies to enhance their efficacy CPI-1205 is a potent, reversible, SAM-competitive small molecule inhibitor of EZH2 (11) CPI-1205 was well tolerated in a phase 1 clinical trial in patients with non-Hodgkin’s lymphoma (12), and is currently being evaluated in a phase 1b/2 trial in patients with mCRPC (ProSTAR, NCT03480646) EZH2 EED SUZ12 Enhancer TSS N N O H N HN O O F F F CPI-1205 H3K27me3 EZH2 inhibition remains effective in cell lines with enhanced AR signaling CPI-1205 combines with ARS inhibitors resulting in synergistic or additive cell growth inhibition Transcriptomic analysis reveals CPI-1205 restricts PC growth in manner distinct from enzalutamide, yet enhances enza effect when combined EZH2 expression is high and polycomb repression signature low in mCRPC Polycomb repression signature Yu, et al. (2007) Cancer Research Comparison of RNA expression from TCGA vs. mCRPC datasets TCGA mCRPC 0 2 4 6 8 **** EZH2 TCGA mCRPC 0 5 10 15 Adj Log 2 FPKM **** AR TCGA mCRPC -6000 -4000 -2000 0 2000 4000 ssGSEA projection **** Signature Comparison of RNA expression data from the TCGA PRAD cohort (majority non mCRPC) to Robinson, et al. (2015) Cell and Robinson, et al. (2017) Nature mCRPC RNA sequencing data 0.01 0.1 1 10 0 25 50 75 100 LNCaP CPI-1205 (μM) Relative cell viability (%) D7 D14 D21 D28 0.01 0.1 1 10 0 25 50 75 100 125 PC3 CPI-1205 (μM) Relative cell viability (%) D7 D14 D21 D28 sensitive EZH2 inhibition effective in cell lines dependent on AR signaling 0.126 GI50: 0.336 0.01 0.1 1 10 0 25 50 75 100 CPI-1205 (μM) Relative cell viability (%) - doxycycline + doxycycline 0 50 100 150 200 250 300 0.01 0.1 1 10 LNCaP CSS +1 nM R1881 enzalutamide (μM) Relative cell viability (%) D7 D14 D21 D28 0.01 0.1 1 10 0 25 50 75 100 CPI-1205 (μM) Relative cell viability (%) 0.196 0.069 GI50: LNCaP cell line engineered with doxycycline-inducible AR overexpression transgene; cell line grown in full serum for viability assays with serial cell re-seeding and drug re-application; day 14 data shown in figure A A B LNCaP cells grown in charcoal-stripped serum before stimulation with 1 nM R1881 to drive proliferation through AR; then grown in the presence of a titration of both CPI-1205 and enzalutamide, subjected to serial cell re-seeding and drug re-application B Specificity of CPI-1205 mediated cell growth inhibition EED Day: 0 4 dox - 22Rv1 (EED KO +EED) dox + 6 8 10 12 14 15 17 18 EZH2 AR ACTB H3K27me3 H3K27ac H3 0.01 0.1 1 10 0 25 50 75 100 125 LNCaP dox- CPI-1205 (μM) Relative cell viability (%) 0.01 0.1 1 10 0 25 50 75 100 125 LNCaP dox+ CPI-1205 (μM) doxycycline: - + LNCaP-EV 0.344 0.063 LNCaP-EZH2 (WT) 0.213 0.059 LNCaP-EZH2 (Y111L) 0.281 3.027 GI50 ( μ M) Day 12 Secondary mutations in EZH2, such as Y111L, identified in vitro using cellular models grown in the presence of EZH2 SMIs confer resistance to the inhibitor (Gibaja, et al. [2015] Oncogene) LNCaP cell line engineered to express doxycycline-inducible expression of EZH2 WT or Y111L, then treated with CPI-1205 for 12 days Endogenous EED-knockout 22Rv1 cell line generated with CRISPR in the presence of a doxycycline-inducible EED transgene Doxycycline removed on day 0, and cells grown for 14 days before re-application on day 15 10 0.01 0.1 1 0 25 50 75 100 combination CPI-1205 (μM) Relative cell viability (%) +DMSO +0.123 μM enzalutamide +0.37 μM enzalutamide +1.111 μM enzalutamide +3.333 μM enzalutamide +10 μM enzalutamide 0.01 0.1 1 10 0 25 50 75 100 normalized combination CPI-1205 (μM) Relative cell viability (%) Day 14 co-treatment Single-cell RNA-sequencing (LNCaP) “Population” RNA-sequencing Bliss excess volume 0 2 2 2 1 -1 0 0 0 0 2 2 2 1 0 0 0 0 0 4 4 4 3 2 1 0 0 0 10 10 10 6 5 3 1 0 0 22 28 28 18 16 12 5 2 0 0 0 0 0 0 0 0 0 enza CPI-1205 EZH2 inhibition is effective in prostate cancer xenograft models, and combines with enzalutamide for enhanced efficacy in LNCaP model treatment group TGI regressions/ total vehicle 0% 0/8 CPI-1205 35% 0/7 enzalutamide 40% 0/8 CPI-1205 + enzalutamide 65% 2/7 B 7.5 5.0 2.5 0.0 -2.5 -5.0 -10 -5 0 Component 1 Component 2 5 10 Cell Treatment DMSO (day 4) CPI-1205, high (day 4) CPI-1205, low (day 7) CPI-1205, high (day 7) 7.5 5.0 2.5 0.0 -2.5 -5.0 -10 -5 0 Component 1 Component 2 5 10 Cell Cycle Phase G1_S S G2 G2_M M_G1 none LNCaP cells treated for 7 days with low (L) or high (H) dose CPI-1205 or DMSO control; heatmap shows differentially expressed genes (DEGs) for high dose CPI-1205 vs. DMSO comparison, and gene set enrichment analysis results LNCaP dataset compared to 22Rv1 EED knockout cells (or wild type control) with re-expression of EED transgene; heatmap shows intersection of DEGs for LNCaP CPI-1205 vs. DMSO and 22Rv1 EED KO day 8 vs. 0 LNCaP cells treated as above were subjected to single cell RNA-seq (10X Genomics platform); results displayed as Monocle trajectory analysis LNCaP cells treated with CPI-1205, enzalutamide, or combination of both followed by RNA-seq; heatmap shows DEGs for combination vs. DMSO A B C D C A Conclusions D PC cell models dependent on AR signaling are sensitive to EZH2 inhibition resulting in time- and dose-dependent cell growth inhibition, with models engineered to have elevated levels of AR signaling becoming even more sensitive to EZH2 inhibition despite reduced sensitivity to ARS inhibitors CPI-1205 combined with ARS inhibitors, such as enzalutamide or abiraterone, in vitro and in vivo resulting in synergistic or additive cell or tumor growth inhibition, with the ability to inhibit the growth of ARS inhibitor-resistant cell lines Transcriptomic analysis revealed modulation of several specific pathways following EZH2 inhibition or EED knockout, showing modest upregulation of AR pathway genes, genes associated with prostate lineage, and importantly down-regulation of proliferation or cell cycle-associated genes Despite EZH2 inhibition modestly upregulating the AR pathway, when combined, CPI-1205 enhances the anti-androgen effect of enzalutamide while maintaining alteration of prostate lineage genes resulting in an anti-proliferative response that is more potent than using either single agent alone This data supports the rationale for the use of CPI-1205 in combination with ARS inhibitors in mCRPC, and importantly shows that EZH2 inhibition can overcome mechanisms of resistance to ARS inhibitors D7 D14 D21 D28 D7 D14 D21 D28 D7 D14 D21 D28 CPI-1205 >10 >10 >10 >10 enzalutamide >10 >10 >10 >10 0.00 0.24 0.32 0.50 na na na na abiraterone >10 >10 >10 >10 0.00 0.09 0.15 0.05 na na na na CPI-1205 >10 >10 >10 >10 enzalutamide >10 >10 >10 >10 0.00 0.08 0.19 0.23 na na na na abiraterone >10 8.37 7.87 8.33 0.04 0.16 0.34 0.10 na na na na CPI-1205 >10 4.09 0.52 0.26 enzalutamide >10 4.52 0.92 1.01 0.69 3.36 2.80 2.17 na 0.21 0.34 0.51 abiraterone >10 >10 >10 >10 0.25 3.69 3.28 3.36 na na 0.28 0.31 CPI-1205 >10 0.52 0.31 0.23 enzalutamide >10 7.33 6.15 2.53 0.69 0.51 0.76 1.71 na 0.40 0.60 0.43 abiraterone >10 3.54 3.95 4.00 0.32 0.34 0.26 0.67 na 0.42 0.72 0.58 CPI-1205 >10 0.95 0.42 0.20 enzalutamide 1.04 0.18 0.15 0.14 2.93 1.34 0.91 0.71 na 0.38 0.57 0.61 abiraterone >10 5.10 2.06 1.79 1.81 3.05 1.67 1.59 na 0.30 0.61 0.68 CPI-1205 >10 >10 0.20 0.07 enzalutamide >10 >10 3.88 3.86 0.47 3.46 5.25 2.76 na na na na abiraterone >10 >10 >10 9.73 0.06 0.59 1.01 0.84 na na na na CPI-1205 >10 >10 0.65 0.32 enzalutamide 0.95 0.20 0.16 0.17 0.06 0.27 0.42 0.53 na na 0.44 0.56 abiraterone nd nd nd nd nd nd nd nd nd nd nd nd Bliss synergy score combo w/ CPI-1205 Combination index combo w/ CPI-1205 Cell Line Treatment GI50 ( μM) LNCaP +1 nM R1881 LNCaP +60 pM R1881 VCaP 22Rv1 LNCaP DU145 PC3 Cell lines treated with combination of CPI-1205 and either enzalutamide or abiraterone for up to 28 days Cells were split, re-seeded, and fed fresh drug every 7 days Combinatorial activity was determined using both Bliss and Loewe models of synergy, and combination index (CI) determined for the GI50 effect level 1. Béguelin W, et al. Cancer Cell. 2013;23(5):677–692. 2. Bradley WD, et al. Chem. Biol. 2014;21(11):1463–1475. 3. Varambally S, et al. Nature. 2002;419(6907):624–629. 4. Yu J, et al. Cancer Res. 2007;67(22):10657–10663. 5. Weiner AB, et al. Prostate Cancer Prostatic Dis. 2016;19(4):395–397. 6. Scher HI, et al. Clin. Cancer Res. 2005;11(14):5223–5232. 7. Beltran H, et al. Cancer Discov. 2011;1(6):487–495. 8. Ku SY, et al. Science. 2017;355(6320):78–83. 9. Mu P, et al. Science. 2017;355(6320):84–88. 10. Xiao L, et al. Cancer Res. 2018;78(20):5731–5740. 11. Vaswani RG, et al. J. Med. Chem. 2016;59(21):9928-41. 12. Harb W, et al. Ann. Oncol. 2018;29(suppl_3). References 0 2 4 6 8 -6000 -4000 -2000 0 2000 4000 EZH2, AdjLog 2 FPKM ssGSEA projection R 2 =0.51 p<0.0001 TCGA dataset mCRPC dataset Cell lines treated with titration of CPI-1205 for 28 days total with serial cell re-seeding and drug re-application every 7 days vinculin EZH2 (WT) EZH2 (Y111L) dox: - + + + + + - - - - LNCaP FLAG EZH2 H3K27me3 H3 μM CPI-1205: - 1.25 0.3 - 5 A A B B Loewe model 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.00 0.20 0.40 0.60 0.80 1.00 enzalutamide (μM) CPI-1205 (μM) Isobologram for GI50 CI = 0.38 0 5 10 15 20 25 30 0 500 1000 1500 22Rv1 (castrated) treatment day tumor volume (mm 3 ) 0 5 10 0 200 400 600 LNCaP (intact) treatment day tumor volume (mm 3 ) vehicle (PO, BID) CPI-1205 (200 mpk, PO, BID) enzalutamide (10 mpk, PO, QD) combination vehicle (PO, BID) CPI-EZH2i (100 mpk, PO, BID) enzalutamide (10 mpk, PO, QD) Day 5 0 2 4 6 Tumor PD ratio H3K27me3 / H3 (%) Day 28 0 1 2 3 4 8 12 Tumor PD ratio H3K27me3 / H3 (%) treatment group TGI >50%TGI vehicle 0% 0/8 CPI-EZH2i 75% 7/7 enzalutamide 12% 1/8 Y111 EZHi AR expression AR mutation AR-V7 expression ETS fusion CPI-1205 sensitive no no no no no no no no no no yes no yes yes yes yes H875Y yes no yes yes T878A no no yes
Transcript
Page 1: EZH2 inhibition as an effective treatment for metastatic ... · EZH2 inhibition is effective in prostate cancer xenograft models, and combines with enzalutamide for enhanced efficacy

0.1110

LNCaP

22Rv1

VCaP

PC3

DU145

GI50 (µM CPI-1205)

0%

50%

100%

150%

200%

D4 D7 D11D14

Cell

viab

ility

Gene sets enriched :

GO_TISSUE_DEVELOPMENTGO_EPITHELIUM_DEVELOPMENT

GO_REGULATION_OF_CELL_DIFFERENTIATION

HALLMARK_ANDROGEN_RESPONSENELSON_RESPONSE_TO_ANDROGEN_UP

BENPORATH_PROLIFERATION

EZH2 inhibition as an effective treatment for metastatic castration-resistant prostate cancerWilliam D. Bradley¹, John P. McGrath¹, CC Yuan¹, Feng Zhao¹, Priyanka Sawant¹, Charlie Hatton², Xinwei Han², Barbara Bryant², Andrew Conery², and Patrick Trojer¹

Translational Sciences¹, Functional Genomics², Constellation Pharmaceuticals, 215 First Street, Cambridge, MA, USA 02142

Introduction

0

50

100

150

200

0.01 0.1 1 10

LNCaP-AR

enzalutamide (µM)

Rel

ativ

ece

llvi

abilit

y(%

) - doxycycline+ doxycycline

AR

ACTB

LNCaP-EV LNCaP-AR- 500 - 10 50 100 500 1000 ng/ml doxycycline

DayDox

22Rv1 EED KO

0+ + +- - -

4 8 14 15 18 0+ -

8

22Rv1 EED WT

- HL

LNCaP+CPI-1205

- HL

LNCaP+CPI-1205

0 2row z-score

-2

√ NELSON_RESPONSE_TO_ANDROGEN_UP √√ HALLMARK_ANDROGEN_RESPONSE √√ WONG_ADULT_TISSUE_STEM_MODULE √

SCHAEFFER_PROSTATE_DEVELOPMENT_48HR_UP √NUYTTEN_EZH2_TARGETS_UP √

√ HALLMARK_E2F_TARGETSGO_REGULATION_OF_CELL_PROLIFERATION √NUYTTEN_EZH2_TARGETS_DN √

Gene sets enriched:

enzalutamideCPI-1205

+- +

+

LNCaP

-- +

-

Enhancement ofenzalutamide effect

Maintenance ofenzalutamide effect

Maintenance ofCPI-1205 effect

EZH2 is the catalytic subunit of the PRC2 complex, and methylates histone H3 on lysine 27, resulting in suppression of gene transcription EZH2 mutations and increased expression are often observed in cancer, leading to repression of genes associated with apoptosis and differentiation (1-2)Over-expression in metastatic castration-resistant prostate cancer (mCRPC [3]), and decreased expression of target genes correlate with poor prognosis (4)While accounting for a small percentage of the overall disease incidence (5), mCRPC is very aggressive with high morbidity (6)Many mCRPC tumors remain dependent on androgen receptor signaling (ARS), but eventually develop resistance to ARS inhibitors, such as enzalutamide and abiraterone acetateRecent studies have linked EZH2 inhibition to re-sensitization to ARS inhibitors in neuroendocrine prostate cancer (NEPC) (7-9), and other PC models (10)Given this unmet need, and the link between EZH2 and advanced prostate cancer, we sought to determine if EZH2 inhibition can be utilized as an effective treatment and/or be combined with ARS inhibitor therapies to enhance their efficacyCPI-1205 is a potent, reversible, SAM-competitive small molecule inhibitor of EZH2 (11)CPI-1205 was well tolerated in a phase 1 clinical trial in patients with non-Hodgkin’s lymphoma (12), and is currently being evaluated in a phase 1b/2 trial in patients with mCRPC (ProSTAR, NCT03480646)

EZH2EED

SUZ12

Enhancer TSS

N

N

O

HNHN

O

O

FFF

CPI-1205

H3K27me3

EZH2 inhibition remains effective in cell lines with enhanced AR signaling

CPI-1205 combines with ARS inhibitors resulting insynergistic or additive cell growth inhibition

Transcriptomic analysis reveals CPI-1205 restricts PC growth in mannerdistinct from enzalutamide, yet enhances enza effect when combined

●●

EZH2 expression is high and polycomb repression signature low in mCRPC

Polycomb repression signatureYu, et al. (2007) Cancer Research

Comparison of RNA expression from TCGA vs. mCRPC datasets

TCG

Am

CR

PC

0

2

4

6

8 ****EZH2

TCG

Am

CR

PC

0

5

10

15

Adj L

og2

FPK

M

****AR

TCG

Am

CR

PC

-6000

-4000

-2000

0

2000

4000

ssG

SEA

pro

ject

ion

****Signature

Comparison of RNA expression data from the TCGA PRAD cohort (majority non mCRPC) to Robinson, et al. (2015) Cell and Robinson, et al. (2017) Nature mCRPC RNA sequencing data

0.01 0.1 1 100

25

50

75

100

LNCaP

CPI-1205 (µM)Rel

ativ

e ce

ll vi

abili

ty (%

)

D7D14D21D28

0.01 0.1 1 100

25

50

75

100

125PC3

CPI-1205 (µM)

Rel

ativ

e ce

ll vi

abili

ty (%

)

D7 D14 D21 D28sensitive

EZH2 inhibition effective in cell lines dependent on AR signaling

0.126

GI50:0.336

0.01 0.1 1 100

25

50

75

100

CPI-1205 (µM)

Rel

ativ

ece

llvi

abilit

y(%

)

- doxycycline+ doxycycline

0

50

100

150

200

250

300

0.01 0.1 1 10

LNCaPCSS +1 nM R1881

enzalutamide (µM)

Rel

ativ

ece

llvi

abilit

y(%

) D7D14D21D28

0.01 0.1 1 100

25

50

75

100

CPI-1205 (µM)

Rel

ativ

ece

llvi

abilit

y(%

)

0.1960.069

GI50:

LNCaP cell line engineered with doxycycline-inducible AR overexpression transgene; cell line grown in full serum for viability assays with serial cell re-seeding and drug re-application; day 14 data shown in figure

A

A

B

LNCaP cells grown in charcoal-stripped serum before stimulation with 1 nM R1881 to drive proliferation through AR; then grown in the presence of a titration of both CPI-1205 and enzalutamide, subjected to serial cell re-seeding and drug re-application

B

Specificity of CPI-1205 mediated cell growth inhibition

EED

Day: 0 4dox -

22Rv1 (EED KO +EED)dox +

6 8 10 12 14 15 17 18

EZH2AR

ACTB

H3K27me3

H3K27ac

H3

0.01 0.1 1 100

25

50

75

100

125

LNCaPdox-

CPI-1205 (µM)

Rel

ativ

e ce

ll vi

abilit

y (%

)

0.01 0.1 1 100

25

50

75

100

125

LNCaPdox+

CPI-1205 (µM)

doxycycline: - +LNCaP-EV 0.344 0.063

LNCaP-EZH2 (WT) 0.213 0.059

LNCaP-EZH2 (Y111L) 0.281 3.027

GI50 (μM)

Day 12Secondary mutations in EZH2, such as Y111L, identified in vitro using cellular models grown in the presence of EZH2 SMIs confer resistance to the inhibitor (Gibaja, et al. [2015] Oncogene)

LNCaP cell line engineered to express doxycycline-inducible expression of EZH2 WT or Y111L, then treated with CPI-1205 for 12 days

Endogenous EED-knockout 22Rv1 cell line generated with CRISPR in the presence of a doxycycline-inducible EED transgene

Doxycycline removed on day 0, and cells grown for 14 days before re-application on day 15

100.01 0.1 10

25

50

75

100

combination

CPI-1205 (µM)

Rel

ativ

e ce

ll vi

abilit

y (%

) +DMSO+0.123 µM enzalutamide+0.37 µM enzalutamide+1.111 µM enzalutamide+3.333 µM enzalutamide+10 µM enzalutamide

0.01 0.1 1 100

25

50

75

100

normalized combination

CPI-1205 (µM)

Rel

ativ

e ce

ll vi

abilit

y (%

)

Day 14 co-treatment

Single-cell RNA-sequencing (LNCaP)

“Population” RNA-sequencingBliss excess volume

0 2 2 2 1 -1 0 0 0

0 2 2 2 1 0 0 0 0

0 4 4 4 3 2 1 0 0

0 10 10 10 6 5 3 1 0

0 22 28 28 18 16 12 5 2

0 0 0 0 0 0 0 0 0

enza

CPI-1205

EZH2 inhibition is effective in prostate cancer xenograft models,and combines with enzalutamide for enhanced efficacy in LNCaP model

treatment group TGI regressions/ total

vehicle 0% 0/8

CPI-1205 35% 0/7

enzalutamide 40% 0/8

CPI-1205 + enzalutamide 65% 2/7

B

7.5

5.0

2.5

0.0

-2.5

-5.0-10 -5 0

Component 1

Com

pone

nt 2

5 10

Cell Treatment

DMSO (day 4)CPI-1205, high (day 4)CPI-1205, low (day 7)CPI-1205, high (day 7)

7.5

5.0

2.5

0.0

-2.5

-5.0-10 -5 0

Component 1

Com

pone

nt 2

5 10

Cell Cycle Phase

G1_SSG2G2_MM_G1none

LNCaP cells treated for 7 days with low (L) or high (H) dose CPI-1205 or DMSO control; heatmap shows differentially expressed genes (DEGs) for high dose CPI-1205 vs. DMSO comparison, and gene set enrichment analysis results

LNCaP dataset compared to 22Rv1 EED knockout cells (or wild type control) with re-expression of EED transgene; heatmap shows intersection of DEGs for LNCaP CPI-1205 vs. DMSO and 22Rv1 EED KO day 8 vs. 0

LNCaP cells treated as above were subjected to single cell RNA-seq (10X Genomics platform); results displayed as Monocle trajectory analysis

LNCaP cells treated with CPI-1205, enzalutamide, or combination of both followed by RNA-seq; heatmap shows DEGs for combination vs. DMSO

A

B

C

DC

A

Conclusions

D

PC cell models dependent on AR signaling are sensitive to EZH2 inhibition resulting in time- and dose-dependent cell growth inhibition, with models engineered to have elevated levels of AR signaling becoming even more sensitive to EZH2 inhibition despite reduced sensitivity to ARS inhibitorsCPI-1205 combined with ARS inhibitors, such as enzalutamide or abiraterone, in vitro and in vivo resulting in synergistic or additive cell or tumor growth inhibition, with the ability to inhibit the growth of ARS inhibitor-resistant cell linesTranscriptomic analysis revealed modulation of several specific pathways following EZH2 inhibition or EED knockout, showing modest upregulation of AR pathway genes, genes associated with prostate lineage, and importantly down-regulation of proliferation or cell cycle-associated genesDespite EZH2 inhibition modestly upregulating the AR pathway, when combined, CPI-1205 enhances the anti-androgen effect of enzalutamide while maintaining alteration of prostate lineage genes resulting in an anti-proliferative response that is more potent than using either single agent aloneThis data supports the rationale for the use of CPI-1205 in combination with ARS inhibitors in mCRPC, and importantly shows that EZH2 inhibition can overcome mechanisms of resistance to ARS inhibitors

D7 D14 D21 D28 D7 D14 D21 D28 D7 D14 D21 D28CPI-1205 >10 >10 >10 >10

enzalutamide >10 >10 >10 >10 0.00 0.24 0.32 0.50 na na na naabiraterone >10 >10 >10 >10 0.00 0.09 0.15 0.05 na na na na

CPI-1205 >10 >10 >10 >10enzalutamide >10 >10 >10 >10 0.00 0.08 0.19 0.23 na na na na

abiraterone >10 8.37 7.87 8.33 0.04 0.16 0.34 0.10 na na na naCPI-1205 >10 4.09 0.52 0.26

enzalutamide >10 4.52 0.92 1.01 0.69 3.36 2.80 2.17 na 0.21 0.34 0.51abiraterone >10 >10 >10 >10 0.25 3.69 3.28 3.36 na na 0.28 0.31

CPI-1205 >10 0.52 0.31 0.23enzalutamide >10 7.33 6.15 2.53 0.69 0.51 0.76 1.71 na 0.40 0.60 0.43

abiraterone >10 3.54 3.95 4.00 0.32 0.34 0.26 0.67 na 0.42 0.72 0.58CPI-1205 >10 0.95 0.42 0.20

enzalutamide 1.04 0.18 0.15 0.14 2.93 1.34 0.91 0.71 na 0.38 0.57 0.61abiraterone >10 5.10 2.06 1.79 1.81 3.05 1.67 1.59 na 0.30 0.61 0.68

CPI-1205 >10 >10 0.20 0.07enzalutamide >10 >10 3.88 3.86 0.47 3.46 5.25 2.76 na na na na

abiraterone >10 >10 >10 9.73 0.06 0.59 1.01 0.84 na na na naCPI-1205 >10 >10 0.65 0.32

enzalutamide 0.95 0.20 0.16 0.17 0.06 0.27 0.42 0.53 na na 0.44 0.56abiraterone nd nd nd nd nd nd nd nd nd nd nd nd

Bliss synergy scorecombo w/ CPI-1205

Combination indexcombo w/ CPI-1205Cell Line Treatment GI50 (μM)

LNCaP +1 nM R1881

LNCaP +60 pM R1881

VCaP

22Rv1

LNCaP

DU145

PC3

Cell lines treated with combination of CPI-1205 and either enzalutamide or abiraterone for up to 28 days

Cells were split, re-seeded, and fed fresh drug every 7 days

Combinatorial activity was determined using both Bliss and Loewe models of synergy, and combination index (CI) determined for the GI50 effect level

1. Béguelin W, et al. Cancer Cell. 2013;23(5):677–692. 2. Bradley WD, et al. Chem. Biol. 2014;21(11):1463–1475. 3. Varambally S, et al. Nature. 2002;419(6907):624–629. 4. Yu J, et al. Cancer Res. 2007;67(22):10657–10663. 5. Weiner AB, et al. Prostate Cancer Prostatic Dis. 2016;19(4):395–397. 6. Scher HI, et al. Clin. Cancer Res. 2005;11(14):5223–5232.

7. Beltran H, et al. Cancer Discov. 2011;1(6):487–495. 8. Ku SY, et al. Science. 2017;355(6320):78–83. 9. Mu P, et al. Science. 2017;355(6320):84–88. 10. Xiao L, et al. Cancer Res. 2018;78(20):5731–5740. 11. Vaswani RG, et al. J. Med. Chem. 2016;59(21):9928-41. 12. Harb W, et al. Ann. Oncol. 2018;29(suppl_3).

References

0 2 4 6 8-6000

-4000

-2000

0

2000

4000

EZH2, AdjLog2FPKM

ssG

SEA

proj

ectio

n

R2=0.51p<0.0001

TCGA datasetmCRPC dataset

Cell lines treated with titration of CPI-1205 for 28 days total with serial cell re-seeding and drug re-application every 7 days

vinculin

EZH2 (WT) EZH2 (Y111L)

dox: - ++ + + +- - - -

LNCaP

FLAGEZH2

H3K27me3

H3

μM CPI-1205: - 1.250.3 - 5

A

A

B

B

Loewe model

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.00 0.20 0.40 0.60 0.80 1.00

enza

luta

mid

e (µ

M)

CPI-1205 (µM)

Isobologram for GI50CI = 0.38

0 5 10 15 20 25 300

500

1000

1500

22Rv1 (castrated)

treatment day

tum

or v

olum

e (m

m3 )

0 5 100

200

400

600

LNCaP (intact)

treatment day

tum

or v

olum

e (m

m3 )

vehicle (PO, BID)CPI-1205 (200 mpk, PO, BID)enzalutamide (10 mpk, PO, QD)combination

vehicle (PO, BID)CPI-EZH2i (100 mpk, PO, BID)enzalutamide (10 mpk, PO, QD)

Day 50

2

4

6Tumor PD

ratio

H3K

27m

e3 /

H3

(%)

Day 280

1

2

348

12

Tumor PD

ratio

H3K

27m

e3 /

H3

(%)

treatment group TGI >50%TGI

vehicle 0% 0/8

CPI-EZH2i 75% 7/7

enzalutamide 12% 1/8

Y111EZHi

AR expre

ssion

AR mutat

ion

AR-V7 exp

ressio

n

ETS fusio

n

CPI-120

5 sen

sitive

no no no no no

no no no no no

yes no yes yes yes

yes H875Y yes no yes

yes T878A no no yes

Recommended