P-MUC1C-101: A Stem Cell Memory CAR-T Therapy for Epithelial-Derived Solid Tumors

Devon J. Shedlock, PhDVP, Preclinical Development

September 5th, 2018

piggyBac™ Enables Multiple Differentiated CAR-T Product Attributes

• Unprecedented cargo capacity (>30X lentivirus) – three-in-one transgene and possibility of multiple CAR or TCR molecules

• Creates highly desirable T Stem Cell Memory (Tscm) Phenotype

• Non-viral delivery system – non-oncogenic and non-mutagenic

• High insertion efficiency and stable transgene expression

• Faster to clinic with lower cost than viral methods

• Substantial IP portfolio with no dominant or competing IP

1

piggyBac™ is a superior DNA delivery system for developing CAR-T and other gene therapy products

GENOMIC DNAITR ITR

“Paste”

CARGO

ITR ITR

“Cut”

CARGO

piggyBacTM

Transposase

ITR ITRCARGO

piggyBacTM

Transposon

+

piggyBac™ Unmatched Cargo Capacity Increases Optionality

2

piggyBac™ effectively delivers multiple full-length CARs in single transposon system

1

2

4

# Full-length CARs* Function (Killing)

* Plus selection gene and marker gene

BCMA

3

BCMA PSMA

BCMA PSMA CD19

BCMA PSMA CD19 GD2

Massive piggyBac™ Cargo Capacity Allows for Delivery of Three-In-One Transgene for CAR-T products

3

CAR-T MOLECULE

Superior binding molecule

• Molecule (Centyrin, VH, scFv, etc…) with high-specificity binding

• Fully human and not susceptible to tonic signaling

1

POSITIVE SELECTION

Drug resistance gene permits positive selection

• ~100% of T-cells in final product express the CAR molecule

• Predicted to result in better therapeutic index

2

SAFETY SWITCH

Incorporates proprietary safety switch

• Rapid, dose-dependent elimination of engineered T-cells if needed

• Management of Cytokine Release Syndrome (CRS) or other AEs

3

TTAA

ITR

Insulator

Promoter Safety Switch CAR Molecule

Selection Gene

Poly(A)

TTAA

Insulator

ITR

Poseida CAR-T Products Comprised of Highly Favorable Stem Cell Memory T Cells

• Ability to develop product with high percentage of Tscmcells is a distinct competitive advantage

• piggyBac™ preferentially transposes in Tscm cells

⎼ Tscm cells persist and live longer than effector cells

⎼ Tscm cells can produce potentially unlimited effectors cells

• Tscm-rich product should lead to better engraftment and better duration of response with the potential for re-response

• Lentivirus-produced products have not achieved high Tscmpublished percentages ranging from less than 1% to ~14%

4

Tscm phenotype should increase duration of response and allow for relapse control without re-administration

TEFF

TEM TCM

TSCM

CD62L

CD

45

RA

Teff Tscm

TcmTem

Tscm May Be Key to Potent and Durable Responses

5

Gattinoni et al. (2009) Nat Med; Hinrichs et al. (2009) PNAS; Hinrichs et al (2011) Blood; Gattinoni et al. (2011) Nat Med; Lugli et al. (2013) JCI; Klebanoff et al

(2016) JCI; Sukumar et al (2016) Cell Met; Sabatino et al. (2016) Blood;

• Correlates with clinical response

⎼ Melenhorst J. et al., UPenn (2017) 20th ASGCT

⎼ Basu et al., Adaptimmune (2017) CAR-TCR Summit

⎼ TCM: Larson, Juno (2018) AACR

• Without prior fractionation of TN/TCM, Akt inhibitors

and shortened process are mostly successful in

increasing TCM during viral manufacture

• “The extreme longevity, the robust proliferative

potential and the capacity to reconstitute a wide-

ranging diversity of the T cell compartment make

the Tscm cell type an ideal cell population to employ

in adoptive immunotherapy”

Poseida CAR-T Products Comprised of Highly Favorable Stem Cell Memory T Cells

6

Adapted from Gattinoni et al. (2017) Nat. Med.

CD62L

CD

45

RA

Teff Tscm

TcmTem

P-BCMA-101

+-++++++++++-

Naïve/TscmTcmTemTeff

P-BCMA-1010 10

310

410

5

Perforin

CD

45R

A85.8%0.5%

13.1%0.6%

CD62L

Poseida CAR-T Comprised of Highly Favorable Tscm

piggyBac™ and Lentivirus Modify Different T Cell Subsets

0

2 0

4 0

6 0

% T

ra

ns

du

ce

d C D 4 + T c e lls

C D 8 + T c e lls

L e n tiv iru s

Cell graphics adapted from Henning et al., Nat. Rev. Immunol., 2018

0

2 0

4 0

6 0

%

Tr

an

sp

os

ed

p i g g y B a c

T M

7

We purified donor cells to these T-cell subsets and then performed optimized piggyBac™ or optimized lentivirus manufacturing on each subset

piggyBac™ preferentially transposes early Tscm cells, while lentivirus prefers differentiated T cells

Poseida CAR-Tscm Unprecedented Preclinical Efficacy

8

P-BCMA-101: Liquid tumor (MM.1S) disseminated IV implantation in NSG mice

P-PSMA-101: Solid tumor (LNCaP) SC implantation in NSG mice

0 5 1 0 1 5 2 0 2 5 3 0 3 5 4 0 4 5 5 0 5 5 6 0

0

5 0 0

1 , 0 0 0

1 , 5 0 0

2 , 0 0 0

2 , 5 0 0

3 , 0 0 0

P r o s t a t e C a n c e r ( L N C a P ) S . C .

D a y s p o s t C A R - T A d m i n i s t r a t i o n

Tu

mo

r

Bu

rd

en

(

Ca

li

pe

r

mm

3)

T c e l l s ( N o C A R ; 4 e 6 )

P - B C M A - 1 0 1 ( 4 e 6 )

P - P S M A - 1 0 1 ( 4 e 6 )

C l i n c i a l P S M A s c F v C A R ( 4 e 6 )

L O D

- 1 4 - 7 0 7 1 4 2 1 2 8 3 5 4 2 4 9 5 6 6 3 7 0 7 7 8 4 9 1

1 05

1 06

1 07

1 08

1 09

1 01 0

1 01 1

D a y s P o s t C A R - T A d m i n i s t r a t i o n

Tu

mo

r

Bu

rd

en

(

BL

I

To

ta

l

Fl

ux

)

L O D

T u m o r O n l y ( n = 1 0 )

P - B C M A - 1 0 1 ( 1 2 e 6 ( n = 2 0 ) )

P - B C M A - 1 0 1 ( 4 e 6 ( n = 1 9 ^ ) )

M u l t i p l e M y e l o m a ( M M . 1 S ) I . V .

9

• Design⎼ Single dose, 3+3 dose escalation + selected expansion⎼ Adult subjects with relapsed or refractory MM

• Key Objectives⎼ Safety, maximum tolerated dose and/or optimal dose⎼ Anti-myeloma efficacy⎼ Expansion and functional persistence of the CARTyrin-T cells

Poseida P-BCMA-101: Phase I Study Ongoing

P-MUC1C-101Autologous CAR-T Therapy for Multiple Solid Tumors

Mucin-1 (MUC1) is highly expressed in most

epithelial-derived cancers

• MUC1 normally expressed on apical surface of

epithelium

• Aberrant form of MUC1 is expressed on

cancer cells, which is specifically recognized by

our binder

• Target for multiple immunotherapies

(e.g. cancer vaccines, antibody therapies)

11

Tumor TypeMUC1

ExpressionNumber

of TissuesReference

Breast 91% 1,447 Rakha et al (2005)

NSCLC 99% 231 Merck Serono. Data on file

RCCa 84% 133 Langner et al (2004)

Colorectal 81% 243 Baldus et al (2002)

Ovarian 83% 63 Chauhan et al (2006)

H&N SCCa 82% 29 Croce et al (2001)

Nasopharyngeal 100% 38 Zhong et al (1993)

Gastric 77% 136 Utsunomiya et al (1998)

Prostate 79% 89 DeNardo et al (2005)

Pancreatic 81% 53 Qu et al (2004)

Mesothelioma 75% 20 Saad et al (2005)

Multiple myeloma 59% 125 Cloosen et al (2006)

Esophageal 32% 53 Kijima et al (2001)

MUC1 Target Has Broad Pan-Tumor Potential

Lambrechts et al

MUC1 is a Highly Complex Protein

Sousa et al 2016

• Single pass Type I transmembrane protein

⎼ N-terminal subunit (MUC1-N) and C-terminal subunit (MUC1-C) form stable heterodimeric complex

• Canonical isoform 1 is a 1,255aa protein

⎼ 42 20aa repeats

• MUC1 is highly polymorphic; various alleles change number of tandem repeats in N’

⎼ >90 isoforms; many with 20-200 VNTR repeats

12

MUC1 is a Highly Complex Protein

Sousa et al 2016

Roulois et al 2013

• MUC1 confined to apical surface of normal epithelial cells

⎼ Hyperglycosylated on normal cells

• Polarity is lost in tumor cells

⎼ Hypoglycosylated on tumor cells; branches seem highly

immunogenic

13

MUC1 can be cleaved by proteases

• MUC1 remains heterodimeric under normal growth conditions

• ECD is cleaved (ADAM 17/TACE and MMP-MT1) to generate shorter membrane-associated peptide fragments (MUC1-C)

⎼ MUC1* and MUC1-CTF15

• Cleaved MUC1-N is shed from cell and triggers inflammation

⎼ Soluble MUC1-N is target for MUC1-N-specific

immunotherapies and may limit their efficacy

14

MUC1 is a Highly Complex Protein

Nath et al 2014

MUC1 Comprises Various Epitopes for CAR Targeting

• Hypoglycosylated branches on tumor cells are highly immunogenic

⎼ Tn-specific binders recognize cancer-associated Tn glycoforms occurring in VNTR (Posey et al, 2016)

• MUC1-C specific binders

⎼ Target epitopes in region proximal to cell membrane

• May be retained post-cleavage

⎼ Are difficult to produce and currently more rare

Tn CARs

MUC1-C CARs

Adapted from Nath et al 201415

16

Many MUC1 Isoforms, But Not All Comprise Tn Epitopes

Exon1 Exon2 Exon3

TnEpitopes

Exon3 Exon4 Exon5 Exon6 Exon7 Exon8

TMMUC1-C Epitopes

Screening CARs Recognizing MUC1

17

Several CAR candidates exhibited antitumor activity against a MUC1+ cancer cell line

Co

ntr

ol

P C

AR

T C

AR

F C

AR

Y C

AR

S C

AR

A C

AR

G C

AR

E C

AR

B C

AR

K C

AR

0

5

1 0

1 5

2 0

2 5

3 0

3 5

4 0

% C

D1

07

a+

R a ji (M U C 1 -)

R P M I8 2 2 6 (M U C 1 + )

We constructed multiple MUC1-C

CARs and tested:

• mRNA CAR Expression in primary

human pan T cells

⎼ Confirmed surface expression of all

• Evaluated antitumor activity against

a MUC1+ cancer cell line

• A few CARs (A,G, and E) expressed

MUC1-specific antitumor activity

Screening CARs Recognizing MUC1

18

G MUC1 CAR exhibited strong antitumor activity against multiple MUC1+ cancer cell lines

We assessed mRNA CAR activity against MUC1 (full-length Isoform 10):

A C

AR

G C

AR

E C

AR

0

2 0

4 0

6 0

8 0

1 0 0

CD

10

7a

+%

e R a ji (F L M U C 1 -Is o 1 0 )

R a ji (M U C 1 -)

N o ta rg e t

G

C

A

R

T

n

C

A

R

M

o

c

k

0

1 0

2 0

3 0

4 0

5 0

6 0

7 0

8 0

9 0

CD

10

7a

+%

M D A - M B - 4 6 8

M C F - 7

R a j i ( M U C 1 - )

N o t a r g e t

We compared mRNA MUC1 (MUC1-C vs TN) CAR activity against several breast cell lines:

MUC1 is Expressed on Numerous Cancers with High Unmet Need

19

Poseida MUC1-C CAR mediates robust anti-tumor activity against multiple tumor lines

CA

R-T

on

ly

BxP

C-3

Pan

c-1

PL

45

SW

1990

Cap

an

-2

Hs 7

66T

T-4

7D

MC

F-7

(S

igm

a)

MC

F-7

(A

TC

C)

MD

A-M

B-4

68

BT

-20

ZR

-75-1

MS

TO

-211H

NC

I-H

226

NC

I-H

441

NC

I-H

1975

A549

NC

I-H

1299

NC

I-H

596

SN

U-C

1

SW

48

LS

123

Hela

K562

RP

MI 8226

U266B

1

AR

P-1

MM

.1S

AG

12211

AG

02102

AG

11132

Raw

264.7

0

1 0

2 0

3 0

4 0

5 0

6 0

7 0

8 0

9 0

CD

10

7a

+%

N o ta rg e t

B re a s t M e s o th e lio m a

& N S C L C

P a n c re a tic C M L & M M

M U C 1 -C C A R

N o rm a l M u rin e

M

A G 1 2 2 1 1 : F ib ro b la s t - L u n g (M U C 1l o

)

A G 0 2 1 0 2 : F ib ro b la s t - U te ru s e n d o m e tr ia l (M U C 1-)

A G 1 1 1 3 2 : E p ith e lia l - B re a s t o rg a n o id (M U C 1-)

N o rm a l c e ll lin e s

C R &

C e rv ic a l

MUC1 is Expressed on Numerous Cancers with High Unmet Need

20

Poseida MUC1-C CAR mediates robust anti-tumor activity against multiple tumor lines including ovarian

M

o

c

k

(

C

o

n

tr

o

l)

P

-P

S

M

A

-1

0

1

(

C

o

n

tr

o

l)

P

-M

U

C

1

C

-1

0

1

T

n

-C

A

R

0

2 0

4 0

6 0

8 0

1 0 0

%

Cy

to

to

xi

ci

ty O V C A R 3 ( O v a r i a n )

M D A . M B . 4 6 8 ( B r e a s t )

M C F 7 ( B r e a s t )

2 4 h r C y t o t o x i c i t yWe assessed P-MUC1C-101 activity against

ovarian tumor line (OVCAR3):

• Strong cytotoxicity

• CAR-Ts also proliferated and secreted IFNg

• Studies evaluating additional ovarian lines

underway

Poseida piggyBac™ manufacture of P-MUC1C-101

21

piggyBac™ manufacturing process yields high levels of CAR-Tscm

Q1

2

Q2

59

Q3

33

Q4

6

0 103

104

105

CD62L

0

103

104

105

CD

45R

A

Q1

1

Q2

59

Q3

39

Q4

1

0 103

104

105

CD62L

0

103

104

105

CD

45R

A

Q1

1

Q2

58

Q3

40

Q4

1

0 103

104

105

CD62L

0

103

104

105

CD

45R

A

Q1

1

Q2

40

Q3

58

Q4

1

0 103

104

105

CD62L

0

103

104

105

CD

45R

A

Mock P-BCMA-101 P-MUC1C-101 Tn CAR

CD62L

CD

45

RA

Teff Tscm

TcmTem

Evaluation of P-MUC1C-101 in vitro

22

Robust in vitro Tumor Killing by MUC1-C CAR-Tscm

1 : 2 1 : 5 1 : 1 0

- 5 0

- 2 5

0

2 5

5 0

7 5

1 0 0B r e a s t c a n c e r M C F - 7

%

Ki

ll

in

g

1 : 2 1 : 5 1 : 1 0- 5 0

- 2 5

0

2 5

5 0

7 5

1 0 0B r e a s t c a n c e r M D A - M B - 4 6 8

P B C M A 1 0 1

P - M U C 1 C - 1 0 1

T n C A R

M o c k

1 : 2 1 : 5 1 : 1 0

- 7 5

- 5 0

- 2 5

0

2 5

5 0

7 5

1 0 0M u l t i p l e m y e l o m a H 9 2 9

%

Ki

ll

in

g

1 : 2 1 : 5 1 : 1 0

- 7 5

- 5 0

- 2 5

0

2 5

5 0

7 5

1 0 0N o r m a l c e l l l i n e A G 0 2 1 0 2 ( M U C 1

-

)

Evaluation of P-MUC1C-101 in vivo

23

MUC1 CAR-Tscm were evaluated in MCF-7 breast cancer orthotopic tumor model

Day 0: CAR-T Treatment1) Tumor only control (n=8)2) Tumor + P-BCMA-101 control (12e6; n=4)3) Tumor + P-MUC1C-101 (12e6; n=4)4) Tumor + Tn CAR (12e6; n=4)

Days -21

Day -21:

MCF-7.luc 3.5x106

Left FP (n=20)

Caliper

7

BLI ImagingBlood Draws

14 21 28 35 42 490-2

Efficacy of MUC1 CAR-Tscm(12e6) in breast cancer model

• MCF-7.luc tumor implanted into left fat pad 21 days prior to CAR-T treatment

• Also serves as tox model since P-MUC1-101 cross-reacts with murine MUC1

Potent in vivo Efficacy of P-MUC1C-101

24

Tumor elimination in 100% of animals at standard dose in human breast cancer xenograft model

0 7 1 4 2 1 2 8 3 5 4 2 4 9

0

5 0 0

1 , 0 0 0

1 , 5 0 0

2 , 0 0 0

2 , 5 0 0

Tu

mo

r

Vo

lu

me

(

mm

3)

D a y s p o s t T c e l l i n j e c t i o n

*

**

* a n i m a l s u c c u m e d

t o t u m o r

0 7 1 4 2 1 2 8 3 5 4 2 4 9

0

5 0 0

1 , 0 0 0

1 , 5 0 0

2 , 0 0 0

2 , 5 0 0

D a y s p o s t T c e l l i n j e c t i o n

* ** a n i m a l s u c c u m e d

t o t u m o r

0 7 1 4 2 1 2 8 3 5 4 2 4 9

0

5 0 0

1 , 0 0 0

1 , 5 0 0

2 , 0 0 0

2 , 5 0 0

D a y s p o s t T c e l l i n j e c t i o n

0 7 1 4 2 1 2 8 3 5 4 2 4 9

0

5 0 0

1 , 0 0 0

1 , 5 0 0

2 , 0 0 0

2 , 5 0 0

D a y s p o s t T c e l l i n j e c t i o n

*

* a n i m a l s u c c u m e d

t o t u m o r

0 7 1 4 2 1 2 8 3 5 4 2 4 9

0

5 0 0

1 , 0 0 0

1 , 5 0 0

2 , 0 0 0

2 , 5 0 0

c a l i p e r m e a s u r e m e n t s

Tu

mo

r

Vo

lu

me

(

mm

3)

D a y s p o s t T c e l l i n j e c t i o n

T u m o r o n l y c o n t r o l ( n = 8 )

P - B C M A 1 0 1 c o n t r o l ( n = 4 )

P - M U C 1 C - 1 0 1 ( n = 4 )

T n C A R ( n = 4 )

*

*

*

*

*

*

* a n i m a l s u c c u m e d t o t u m o r

Summary: Developing MUC1 CAR-T to Address Multiple Indications

25

Discovery program with compelling preclinical data and multiple development options

• P-MUC1C-101 demonstrated robust antitumor activity against multiple tumor types

• P-MUC1C-101 Tscm mediated rapid, sustained tumor regression and completely eliminated established tumors in an MCF-7 based orthotopic mouse model

• Compelling data suggesting molecule is binding tumor-specific MUC1

• Preclinical data portends broad product potential

AcknowledgmentsEric Ostertag, M.D., Ph.D, CEOMark J. Gergen, J.D., CBO & CFOMatthew Spear, M.D., CMO

Devon J Shedlock, Ph.D., VP of Preclinical Development

Immuno-OncologyJenessa Smith, Ph.D.Xinxin Wang, Ph.D.*Burton Barnett, Ph.D.Stacey Cranert, Ph.D.Christopher Martin, Ph.D.Elvira Khialeeva, Ph.D.Srinivas Rengarajan, M.S.Yenning Tan, M.S.Rebecca Codde