Figure 4. All cell lines (except for SU-DHL-10) were treated with Tazemetostat for 4 days and then cells were normalized and treated with the indicated agent for 3 additional days. Tazemetostat pre-treatment of DLBCL cells lines (both EZH2 mutant and wild-type) can sensitize cells to inhibitors of B-cell activation pathways. The anti-proliferative effect of tazemetostat can be diminished by combination with B-cell activators. * Additivity = when both agents have single agent effect but no significant enhancement of Tazemetostat potency observed. No Effect = when one agent has no single agent effect and causes no enhancement of Tazemetostat potency ^ Tested in 4-day cotreatment model

Figure 4. Combination studies reveal a relationship between B-cell activation and EZH2 inhibition

EZH2 plays a critical role in B-cell maturation and in non-Hodgkin’s lymphoma: Interplay between EZH2 function and B-cell activation

Danielle Johnston1, Dorothy Brach1, Christopher Plescia1, Allison Drew1, Trupti Lingaraj1, Natalie Warholic1, Jesse Smith1, Robert Copeland1, Heike Keilhack1, Elayne Penebre1, Sarah Knutson1, Scott Ribich1, Michael Thomenius1, Alejandra Raimondi1

1Epizyme Inc., 400 Technology Square, Cambridge, MA, USA

The EZH2 inhibitor tazemetostat (EPZ-6438) is emerging as a promising therapeutic agent for the treatment of non-Hodgkin Lymphoma (NHL). A significant body of work has now demonstrated in vitro and in vivo effects of EZH2 inhibition in preclinical models of lymphoma, in addition to objective clinical responses in early human trials. While EZH2 gain-of-function mutations clearly contribute to lymphomagenesis, patients with lymphomas harboring wild-type EZH2 also show responses to tazemetostat. This suggests a broad role of EZH2 in B-cell oncogenesis. Several recent mouse model studies have demonstrated the importance of wild-type EZH2 catalytic activity in the formation of germinal centers in non-diseased lymph nodes, suggesting a central role for EZH2 in B-lymphocyte maturation. These findings indicate that the importance of EZH2 to B-cell lymphoma likely lies in its ability to regulate B-cell differentiation. To understand the relationship between B-cell maturation and sensitivity to EZH2 inhibition, we evaluated changes in maturation markers and cellular proliferation following treatment of diffuse large B-cell lymphoma (DLBCL) cell lines with tazemetostat in combination with modulators of B-cell activation. Consistent with the importance of EZH2 in the regulation of B-cell differentiation, we observed increased expression of B-cell maturation markers in DLBCL cell lines treated with single agent tazemetostat in vitro. Furthermore, we demonstrate that tazemetostat pre-treatment of subsets of DLBCL cells lines (both EZH2 mutant and wild-type) can sensitize cells to inhibitors of B-cell activation pathways, which include glucocorticoids and BTK, MAPK and PI3K pathway inhibitors. Moreover, the anti-proliferative activity of single agent tazemetostat can be diminished or delayed by co-treatment with biological stimulators of B-cell activation including B-cell receptor ligation, CD40L, LPS and BAFF. Importantly, B-cell receptor ligation and co-stimulation agents have little proliferative effects on DLBCL cell lines on their own, suggesting that the protective function of these agents is directly related to the effects of EZH2 inhibition and not a generic stimulation of proliferation. Our findings suggest that EZH2 inhibition initiates a differentiation program that enables lymphoma cells to proceed through the normal processes of B-cell selection, growth regulation and maturation.

Abstract Results

Background

Conclusions • EZH2 regulates maturation of both normal B-cells and lymphoma cells • B-cell differentiation effects are observed in both WT and Mutant EZH2 NHL cell lines in response to

tazemetostat • EZH2 inhibition creates a dependency on B-cell activation signals • Tazemetostat treatment may allow lymphoma cells to attempt to proceed through the normal processes of B-

cell maturation and selection

Figure 1. Clinical effects of tazemetostat suggest that an additional anti-tumor mechanism occurs in vivo that is not observed in vitro. A) Clinical responses observed in response to tazemetostat treatment. B) 11 day proliferation IC50s with tazemetostat for 52 cell lines. Cells lines are grouped by EZH2 mutation status. C) 11 day tazemetostat growth curves for KARPAS-422 (EZH2 Mutant) and SU-DHL-5 (EZH2 Wild-Type) D) OCI-LY19 (EZH2 Wild-Type) xenografts show tumor regression in response to tazemetostat.

Figure 6. EZH2 inhibition is required for CD40L induction of PRDM1 in SUDHL-5

Figure 6. Tazemetostat treatment induces PRDM1 (BLIMP1). A) 4-day treatment of KARPAS-422 (Mutant EZH2) cells but not SU-DHL-5 (Wild-Type EZH2) with tazemetostat (taz) induces PRDM1 expression but 14-day co-treatment of SU-DHL-5 cells with CD40L and tazemetostat (taz) in combination results in a robust induction of PRDM1 expression (bottom). B) 96 hour tazemetostat treatment does not affect IκB degradation in response to CD40L.

www.epizyme.com

Native B cell Germinal Center

Decision Phase

Germinal Center

Dark Zone Light Zone

Marginal Zone

Mantle Zone

Tazemetostat may “open the door”

for differentiation

Dependent on Activation Signals

Terminal Differentiation Apoptosis

Apoptosis

Plasma/Memory Cell

Does Tazemetostat (EPZ-6438) poise lymphoma cells for terminal differentiation?

1) EZH2 inhibitors are emerging as valuable therapeutic agents for the treatment of non-Hodgkin’s lymphoma.

2) The mechanism by which EZH2 inhibitors inhibit lymphoma cell proliferation is still poorly understood.

3) EZH2 is known to be a regulator of B-cell maturation and germinal center formation1,2,3.

4) Lymphomas with gain of function mutations in EZH2 predominantly have a germinal center phenotype4.

5) EZH2 inhibition induces differentiation markers in lymphoma cells.

6) Does EZH2’s role in B-cell maturation contribute to the anti-lymphoma activity of tazemetostat?

Figure 3. EZH2 inhibition induces a gene set that is heavily enriched for plasma cell and memory cell genes in DLBCL cell lines. RNAseq data from KARPAS-422 (EZH2 Mutant), Farage (EZH2 Wild-Type) and SU-DHL-5 (EZH2 Wild-Type) treated with 1 µM tazemetostat for 4 days was analyzed using GSEA (Broad Institute) for the presence of plasma and memory cell gene sets.

Figure 3. EZH2 inhibition induces plasma and memory cell gene sets in cell lines with both wild type and

mutant EZH2.

Karpas 422 mut (0.0018µM) Farage (WT-0.099µM)

NES

= 1

.56

NES

= 1

.8

NES

= 1

.55

NES

= 1

.77

NES

= 1

.59

NES

= 1

.78

SUDHL5 (WT-2µM IC50)

Figure 1. Clinical effects of Tazemetostat suggest that an additional anti-tumor mechanism occurs in vivo beyond what is

observed in vitro.

vehicle 200mg/kg 600mg/kg

GL7

6.3% 2.0% 1.0%

Figure 2. EZH2 inhibition affects B-cell maturation A) Mice treated with EPZ011989 have fewer germinal center B-cells, suggesting enhanced differentiation. Splenic B-cells from treated mice were isolated and analyzed with anti-B220 and anti-GL7 antibody. B) DiscoverX BT Biomap profiling shows an increase in secreted IgG from stimulated PBMC/B-cell co-cultures treated with EPZ07210 which suggests B-cell maturation.

Figure 2. EZH2 inhibition affects normal B-cell maturation

Evaluable Patients (n=15)

DLBCL (n=9)

FL (n=5) MZL (N=1)

CR+PR * 5 3 1

SD 0 1 0

14 NHL patients tested to date: 13 WT + 1 mutant by cobas® EZH2 Mutation Test (in development, Roche Molecular Systems, Inc.) * Cheson Criteria

SU-DHL-5

5 15 30 60 0 5 15 30 60 0 5 15 30 60 0

DMSO 0.1µM

tazemetostat 1µM

tazemetostat 500ng/mL CD40L (min)

IkBα

GAPDH

Figure 5. Loewe volume synergy measurements were determined using CHALICE software. Synergy >2. Additivity = values between -2 and 2. Antagonism < -2. Potency of Tazemetostat is synergistically increased with the addition of Ibrutinib in both EZH2 mutant and WT GCB lymphoma lines (A, B) however there is a much more modest response in the ABC line OCI-LY-3 (C). Potency of Tazemetostat is synergistically increased with the addition of Prednisolone (D, E, and F). Antagonism is observed with the addition of CD40L (which has no single agent effect) in GCB lymphoma lines but not in the ABC line (G, H, I). * See figure 4 for definition of Additivity and No Effect

Figure 5. Primary data from Figure 4

A B

C

PRDM1

GAPDH

14 days SU-DHL-5 4 days

A

B

M i c e t u m o r v o l u m e o f E P Z 0 0 6 4 3 8 e f f i c a c y s t u d y i n t h e

t r e a t m e n t o f O C I - L y 1 9 s . c x e n o g r a f t s

0 7 1 4 2 1

0

5 0 0

1 0 0 0

1 5 0 0

2 0 0 0 V e h i c l e , B I D x 2 0

E P Z 0 0 6 4 3 8 , 1 2 5 m p k , B I D x 2 0

E P Z 0 0 6 4 3 8 , 5 0 0 m p k , B I D x 2 0

D a y s p o s t g r o u p i n g ( D a y )

Tu

mo

r v

olu

me

ch

an

ge

(%

)

OCI-LY19 Xenograft Tumor Volume in response to tazemetostat (EPZ006438)

B85

D

A

B