Poster

The discovery and development of a bispecific T cell engager prodrug targeting ENPP3 for renal cell carcinoma

The discovery and development of a bispecific T cell

engager prodrug targeting ENPP3 for renal cell carcinoma

Li Chen, Lindi Wang, Yingyu Li, Xin Wang, Yang Xin, Mingzhu Shao, Jiangtao Ning, Ziyu Chen, Cheng Luo, Jingjing Nie, Yue Wu, Tinging Yang, Hai Huang,

Miaomiao Song, Yu Liang

Introduction

T cell engager (TCE) antibodies have proven effective for a variety of

hematological malignancies; however, their efficacy in many solid tumors is still

limited due to a number of challenges. One of the leading challenges is the lack of

tumor-specific targets, leading to an increased risk of on-target, off-tumor

toxicity. ENPP3 is exemplary of this with overexpression in 60-95% of renal cell

carcinomas (RCC) and non-trivial levels of expression in normal tissue of the

gastrointestinal tract and respiratory system. One approach to address this

challenge takes advantage of linker and masking technologies that render the TCE

inactive outside of the tumor microenvironment (TME), but active within it. To

develop such a prodrug TCE molecule with improved tumor selectivity, potency,

and safety profile, we first developed a masking peptide specific to our previously

engineered CD3 VHH antibody. Next, we fused this masking peptide to the CD3

VHH arm via a proteolytically-labile linker containing sequence motifs sensitive to

proteases enriched in the TME, and showed that upon cleavage by proteases, the

T cell-activating potency of such a masked CD3 VHH can be effectively recovered.

Finally, with such a CD3 VHH fitted with cleavable linker and masking peptide, we

built a series of prodrug TCE molecules targeting ENPP3 and evaluated their

function through relevant in vitro and in vivo models.

CD3 VHH Masking Peptide Generation

$ %"

! %C

! %"

" %C

" %"

! "

D ( ) * + ) ,-. ,/( ) M1) O 3

! "

(A) From a common camel-derived clone, cyno cross reactive CD3 VHH antibodies were engineered through

humanization, affinity maturation, and PTM site removal to establish a set of “tool box” variants with differing

CD3 binding affinities. (B) Various peptides were screened in a competitive inhibition assay to assess their

ability to interfere with the CD3 VHH antibody’s ability to bind hCD3εδ protein. (C) When joined to the CD3

VHH antibody via linker, Peptide A, but not ctrl peptide H, potently inhibits the CD3 VHH’s ability to bind the

hCD3εδ protein (left) and activate T cells (right).

ProBioCDMO.com

4 5 6 C " M) 7

! " # $% C C $C D( )

! " # $% C C $* H , D- .

! " # $% C C $L M 1

2 P # 4

Unmasked CD3 VHH Binding to CD3εδ

hCD3εδ

Anti-hFc-HRP

CD3 VHH

Blocking Assay by Synthetic Peptide

Masks

! " ^#$

! " ^#%

! " ^#!

! " ^"

! " ^!

! " ^%

! " ^$

C D ( ) *+,- . /

M 1 O 3 " +( 4

! " # ABC " DE

! " # ABC " DF

! " # ABC " DG

! " # ABC " DH

! " # ABC " DI

! " # ABC " DJ

! " # ABC " DK

! " # ABC " DL

! " # ABC " DM

! " # ABC " D1

! " # ABC " D2

! " # ABC " DP

! " # ABC " D4

E " 5 6 7 8 9 :; D8 9 <;

Biotin-Peptide

hCD3εδ

Anti-hFc-HRP

CD3 VHH High

! "

! " # $ % # CD( C)" # *+# , -

. / M 1 O *# 3

! " # A%C " D( FG H I - F. / - D0 ( ( D( %1 2 FP 4

! " # A%C " D5 FG H I - F. / - D0 ( ( D( %1 2 FP 4

. / - D0 ( ( D( %1 2 FP 4

S77" 8" V : ; AD0 ( ( FP 4

Binding Activity of Masked

CD3 VHH to CD3εδ

~1600x

! " # $ % # C'( C)" # *+# , -

. L M )# % 1 $ % # $ % *+2 . 3 -

4 R ^{6 7}

4 R ^{6 U}

4 R ^{6 4}

4 R

: R R R R R

9 R R R R R

7 R R R R R

U R R R R R

4 R R R R R

64 R R R R R

T Cell Activation of Jurkat

Reporter Cells

Conclusions

Using our highly modular CD3 VHH and masking platforms, we were able to optimize an

ENPP3 TCE prodrug with TME-specific activity. With no observed toxicity at doses 100-

fold greater than the unmasked version, our ENPP3 TCE prodrug demonstrates a

potentially large therapeutic index allowing for higher dosing regimens to induce clinical

responses. Further development, including non-human primate pharmacokinetics and

toxicity studies is ongoing to further validate the drug’s therapeutic potential.

Contacts: Yu Liang, Ph.D.; Vice President of Discovery, yu.liang@probiocdmo.com;

Suraj Kachgal, Ph.D.; Head of Global Asset & Technology Licensing, suraj.kachgal@genscript.com

Functionality of Lead ENPP3 x CD3 TCE Prodrug

ENPP3 TCE prodrug does not exhibit TAA independent T cell activation. Human PBMCs were incubated with

the indicated test groups for 48 hr. and (A) T cell activation was assessed via CD25 (left) and CD69 (right)

expression by FACS. (B) Cytokine expression was measured via ELISA for IL-2 (left), TNFα (middle), and IFNγ

(right).

!" ^{# $}

!" ^{# %}

!" ^{# !}

!" ^"

!" ^!

!" ^%

!C""

!"""

C""

D ( F * + F I-. I/( F M1F 2 3

2 4 5M( 6MD T 8 9 M( F MD T $ M: M* + ;;

! " # $! " B C D( ) * + ,

- . / ! DM 1 2P

!" ^{# A}

!" ^{# %}

!" ^{# !}

!" ^"

!" ^!

!" ^%

A""

%""

!""

%C"""

%""""

!C"""

!""""

C"""

D E F * H F I-. ILE F MNF 2 P

2 4 5ME SMD T % C ME F MD T A MU M* H 99:

X F Y . : = H > M? @ A A A MU D ?

B C B ? a #! "

2 . : = H > M? @ A A A MU D ?

b A A c #D d LY H -L*

e Y f g h ! i M. F . 9E M

TAA-Independent T Cell Activation

! " # $! " B C D( ) * + ,

- . / ! DM 1 2P

! " ^{# $}

! " ^{# %}

! " ^{# !}

! " ^"

! " ^!

! " ^%

! C " "

! " " "

C " "

' ( F * + F ,-. ,L( F MNF 2 3

4 5 6 T MN8 9 :; < 3

! " # $! " B C D( ) * + ,

- . / ! DM 1 2P

! " ^{# $}

! " ^{# %}

! " ^{# !}

! " ^"

! " ^!

! " ^%

C " "

$ " "

% " "

! " "

' ( ) * + ) I-. IL( ) M1) 2 3

45 #% M16 7 89 5 3

! " # $! " B C D( ) * + ,

- . / ! DM 1 2P

! " ^{# $}

! " ^{# %}

! " ^{# !}

! " ^"

! " ^!

! " ^%

( " " "

' " " "

C " " "

% " " "

F * + I - + .LM .N* + 23+ 4 5

67 8 9 23: ; <= > 5

TAA-Independent Cytokine Release

! " # $! " B C D( ) * + ,

- . / ! DM 1 2P

! " # A B # C'E CF" # G+# I -

K L M F# B N A B # A B G+2 K P -

4 R ^{6 T}

4 R ^{6 4}

4 R ^R

4 R ⁴

4 R ^T

9 R R R R R

U R R R R R

T R R R R R

W ; < < = G> ! W G6GL # M E N ? B @

A T B ! R I U C a> 4

A T B ! ! 4 U C a> 4

A T B ! ; < b C a> 4

A T B ! P c d C a> 4

A T B ! B e I C a> 4

A T B ! C g T C a> 4

A T B ! 9 e = C a> 4

A T B ! < = d C a> 4

K iE # ? GI B @ FL M

W ; < < = G> ! W G< ' " @ ' L M G6G# " # 6A iB E k E l iB GiF# ? B '

ENPP3 TCE prodrug w/

cleavable linker + RCC CM

Effective release of active TCE

upon cleavage by conditioned

medium derived from RCC

tumor samples. ENPP3 prodrugs

were treated for 2 days with

conditioned medium prepared

by co-culturing with a panel of

tumor samples of 8 RCC

patients, and the release of

active TCE was assessed by

reporter gene assay for T cell

activation.

& !

& "

& #

& $

& B

" !

" "

( !

" D

" !

& D

& !

) * + , -. / , 0-012 * 03 2 4 0

5 / 6 + -7 2 89 W 0-;9 <

ENPP3 TCE prodrug allows for a high-dosing range to

elicit potent antitumor efficacy, better than a leading

clinical benchmark, without noticeable toxicity, in a

KU812 CDX model with human PBMC transplant. The

control, masked TCE with non-cleavable linker did not

show significant tumor growth inhibition.

* p＜0.05, *** p＜0.001, **** p＜0.0001 relative to

Vehicle. ## p＜0.01 relative to Xmab819.

3 4 1 - /+5 - 64 1 0 T1 1 ⁸V

! ! !

! ! ! !

: : : :

Xmab819 analog, 0.1 mg/ kg

51%

Active ENPP3 TCE, 0.062 mg/ kg

73%

ENPP3 TCE Prodrug w/ Cleavable Linker, 0.066 mg/ kg

30%

ENPP3 TCE Prodrug w/ Cleavable Linker, 0.66 mg/ kg

104%

ENPP3 TCE Prodrug w/ Cleavable Linker, 6.6 mg/ kg

109%

ENPP3 TCE Prodrug w/ Uncleavable Linker, 0.066 mg/ kg

ENPP3 TCE Prodrug w/ Uncleavable Linker, 0.66 mg/ kg

ENPP3 TCE Prodrug w/ Uncleavable Linker, 6.6 mg/ kg

31%

Vehicle

D ( ) +" & +3 ; <

5 G =+> 0 ? @A 60

B B

" "

& !

& "

& #

& $

& %

" !

" "

" ! ! !

& $ ! !

& " ! !

% ! !

# ! !

D ( ) * +, - * .+./0 ( .1 0 2 .

3 4 1 - /+5 - 64 1 0 +T1 1 ⁸

! ! !

!! ! !

:: : :

; ;

" "

Effective Activation of Prodrug Upon

Cleavage by RCC Patient Samples

Potent TGI in

No Body Weight Loss

KU812 CDX Model

Development of ENPP3 TCE Prodrug

Mask cleavage restores target cytotoxicity. Peptide A was conjugated to an

ENPP3 TCE (using each of the different affinity CD3 VHHs in a 2+2 format) via an

enzymatically cleavable linker. The resultant prodrug TCEs were treated with the

protease, MTSP-1, and evaluated for T cell activation (A), KU812 tumor cell killing

(B), and the accompanied cytokine release (C) at 24 hr.

5 *D + 6T U 9*) U ::*; - , <,

= L > + , ? U @ *A B C C C *D ) A

= L > + , ? U @ *A B C C C *D ) A *a *E <T U , 9<. L

P + , ? U @ *A B C C C *D ) A

P + , ? U @ *A B C C C *D ) A *a *E <T U , 9<. L

) . L 96. :*D ) A

) . L 96. :*D ) A *a *E <T U , 9<. L

CD3 VHH High Affinity

! " ^{# $}

! " ^{# %}

! " ^"

! " ^%

C "

$ "

% "

' ( ) * + ) ,-. ,L( ) M1) 2 3

4 M5 . -6 + ,M' + TTM8 9 : L:

CD3 VHH Medium Affinity

! " ^{# $}

! " ^{# %}

! " ^"

! " ^%

C "

$ "

% "

' ( ) * + ) ,-. ,L( ) M1) 2 3

4 M5 . -6 + ,M' + TTM8 9 : L:

CD3 VHH Low Affinity

! " ^{# $}

! " ^{# %}

! " ^"

! " ^%

C "

$ "

% "

' ( ) * + ) ,-. ,L( ) M1) 2 3

4 M5 . -6 + ,M' + TTM8 9 : L:

! " ^#$

! " ^#%

! " ^#C

! " ^"

! " ^C

( " "

% " "

' " "

C " "

! " "

F * + , - + .LM .N* + 23+ 4 5

6 T 8 9 23: ; <= > 5

! " ^#$

! " ^#%

! " ^#C

! " ^"

! " ^C

C ' "

C " "

! ' "

! " "

' "

( ) * + I * -.L -M) * 12* 3 4

56 #C 127 8 9: 6 4

! " ^#$

! " ^#%

! " ^#C

! " ^"

! " ^C

% " " "

' " " "

C " " "

! " " "

( ) * + I * -.L -M) * 12* 3 4

56 #$ 127 8 9: 6 4

! " ^#$

! " ^#%

! " ^#C

! " ^"

! " ^C

! ' " "

! " " "

' " "

( F * + I * -.L -MF * N2* 3 4

56 7 8 N29 : ;< = 4

! " ^#$

! " ^#%

! " ^#C

! " ^"

! " ^C

$ " "

% " "

C " "

' ( F * + F ,-. ,L( F MNF 2 3

4 5 6 T MN8 9 :; < 3

! " ^#$

! " ^#%

! " ^#C

! " ^"

! " ^C

C ' "

C " "

! ' "

! " "

' "

( ) * + I * -.L -M) * 12* 3 4

56 #C 127 8 9: 6 4

! " ^#$

! " ^#%

! " ^#C

! " ^"

! " ^C

' " " "

C " " "

! " " "

( ) * + I * -.L -M) * 12* 3 4

56 #$ 127 8 9: 6 4

! " ^#$

! " ^#%

! " ^#C

! " ^"

! " ^C

C " " "

! ' " "

! " " "

' " "

( F * + I * -.L -MF * N2* 3 4

56 7 8 N29 : ;< = 4

! " ^#$

! " ^#%

! " ^#C

! " ^"

! " ^C

' " "

$ " "

% " "

C " "

( F * + , * -.L -MF * N2* 3 4

5 6 T 8 N29 : ;< = 4

! " ^#$

! " ^#%

! " ^#C

! " ^"

! " ^C

' " "

$ " "

% " "

C " "

( ) * + I * -.L -M) * 12* 3 4

56 #C 127 8 9: 6 4

! " ^#$

! " ^#%

! " ^#C

! " ^"

! " ^C

! " " " "

' " " "

$ " " "

% " " "

C " " "

( ) * + I * -.L -M) * 12* 3 4

56 #$ 127 8 9: 6 4

! " ^#$

! " ^#%

! " ^#C

! " ^"

! " ^C

% " " "

' " " "

C " " "

! " " "

( F * + I * -.L -MF * N2* 3 4

56 7 8 N29 : ;< = 4

CD3 VHH High Affinity

CD3 VHH Medium Affinity

CD3 VHH Low Affinity

. L M

α U V . ! : ; G< G! ; T G= > > GH! " # CD" ?-

α @ A V V T G< G! ; T G= > >

α @ A V V T G< G! ; T G= > > PL )# B % DP, E C B

PGGGGGGGG

, a E V N G; )c % C C)" #

CD3 VHH High Affinity

! " # $ % # C'( C)" # *+# , -

. LM

1 2 ^{3 4}

1 2 ^{3 R}

1 2 ^{3 6}

1 2 ²

1 2 ⁶

1 2 ^R

7 2 2 2 2 2

6 2 2 2 2 2

1 2 2 2 2 2

CD3 VHH Medium Affinity

! " # $ % # C'( C)" # *+# , -

. L M

1 2 ³⁴

1 2 ^3R

1 2 ³⁶

1 2 ²

1 2 ⁶

1 2 ^R

7 2 2 2 2 2

6 2 2 2 2 2

1 2 2 2 2 2

CD3 VHH Low Affinity

! " # $ % # C'( C)" # *+# , -

. LM

1 2 ^{3 4}

1 2 ^{3 R}

1 2 ^{3 6}

1 2 ²

1 2 ⁶

1 2 ^R

7 2 2 2 2 2

6 2 2 2 2 2

1 2 2 2 2 2