Characterization of AB154, a Humanized, Non-Depleting α-TIGIT Antibody Undergoing Clinical Evaluation in Subjects with Advanced Solid Tumors

Anderson AE, Lopez A, Udyavar A, Narasappa N, Lee S, DiRenzo D, Zhang K, Singh H, Zhao S, Gerrick K, Park A, Seitz L, Walker N, Walters MJ and Tan JBL.

Arcus Biosciences, Inc., Hayward, CA, USASITC 2019

Poster P260

Introduction

AB154 is a humanized antibody that blocks human TIGIT (T-cell

immunoreceptor with Ig and ITIM domains), an inhibitory receptor expressed on

natural killer (NK) cells, CD8+ T cells, CD4+ T cells and regulatory T cells (Treg).

CD226 (or DNAX Accessory Molecule-1, DNAM-1) is an activating receptor that

competes with TIGIT for shared ligands CD155 (PVR) and CD112 (Nectin-2),

expressed by cancer and antigen-presenting cells. AB154 blocks TIGIT with

sub-nanomolar affinity, thus prevents binding to its ligands and shifts the

immune balance towards a more favorable CD226 interaction. Because AB154

is engineered without FcγR binding function, the immune suppressive TIGIT-

CD155 interaction is blocked with minimal risk of depleting intra-tumoral

antigen-experienced CD8+ T cells. AB154 has the potential to promote

sustained immune activation and tumor clearance, particularly in combination

with other immunotherapies such as AB122 (anti-PD1).

Methods

Results

Gene Expression: Expression of TIGIT, PD-1 (PDCD1), and CD226 on select

tumor types were derived from RNASeq in The Cancer Genome Atlas (TCGA)

database.

Immunohistochemistry (IHC): Anti-CD155 antibody (Cell Signaling

Technology, D8A5G) was used to stain FFPE human tissues. Samples were

deparaffinized according to standard methods and heat-induced epitope

retrieval was performed using sodium citrate. Anti-rabbit HRP and DAB

chromogen were used for detection.

ADCC Assay: AB154 and versions of AB154 modified to restore wild-type (WT)

IgG1 effector function or to display enhanced FcγR binding via Fc mutations

(EEF) were used in antibody-dependent cell cytotoxicity (ADCC) studies.

Identification of Intra-tumoral Antigen-Experienced T cells: Dissociated

tumor samples from head and neck cancer patients were purchased from

Discovery Life Sciences (n = 5). Cells were assessed for viability, surface and

intracellular markers related to T cell lineage, exhaustion, and activation by flow

cytometry.

AB154 Potency in Healthy Volunteers vs. Lung Cancer Patients: Whole

blood was obtained from healthy volunteers (n = 3) and non-small cell lung

carcinoma (NSCLC) patients (n = 3) and assessed by flow cytometry. TIGIT

receptor occupancy was determined using saturating levels of a commercially-

available human α-TIGIT antibody that binds competitively with AB154.

Clinical PK/PD: A Phase 1 dose-escalation study is underway to evaluate

AB154 as a monotherapy and in combination with AB122 (anti-PD1) in subjects

with advanced solid malignancies. Whole blood was assessed by flow

cytometry using saturating levels of a competing anti-TIGIT antibody to

determine receptor occupancy (RO) as well as changes in Ki67 expression.

Conclusions

• TIGIT, PD-1, and CD226 expression are correlated in many tumor types and

are often co-expressed on tumor infiltrating lymphocytes (TILs). CD155 is

also highly expressed by cancer types of interest.

• CD8+ T cells make up the majority of antigen-experienced T cells in

advanced head and neck tumor samples. Antigen experience occurs

alongside markers of immune exhaustion and loss of CD226 expression.

• Antibodies engineered with enhanced effector function (EEF) can facilitate

ADCC on intra-tumoral antigen-specific CD8+ T cells.

• TIGIT receptor density is often comparable or higher on intra-tumoral

antigen-experienced CD8+ T cells than on Treg, highlighting the danger of

targeting these cells with depleting antibodies.

• AB154 potently binds TIGIT in peripheral blood from both healthy volunteers

and NSCLC patients.

• AB154-dosed patients have had complete receptor coverage on all TIGIT-

expressing peripheral leukocytes in this Phase 1 trial (NCT03628677).

Proliferative bursts in CD8+ T cells demonstrate immune engagement.

• A Phase 2 study including the combination of AB154 and AB122 is planned.

TIGIT, PD-1, and CD226 Are Co-Expressed on

Human Tumors

Interested in a career at Arcus? Visit arcusbio.com

Figure 1. TIGIT binds to CD155 and results in decreased activation of the

TIGIT-expressing immune cells. AB154 blockade of TIGIT allows CD155 to bind

CD226, favoring T cell and NK cell activation.

Figure 2. RNASeq data from TCGA reveals high levels of TIGIT and PD-1

(PDCD1) co-expression across many tumor types. CD226 is often expressed in

TIGIThi PD-1hi tumor types.

November 7-10, 2019. National Harbor, Maryland

Figure 6. Patients and healthy donors had similar TIGIT expression in

peripheral blood lymphocytes, including CD8+ and CD8- T cells, NK cells and

NKT cells, as measured using saturating levels of a commercially available α-

TIGIT antibody. Fluorophore-conjugated AB154 was used to directly determine

binding affinity in whole blood, with equipotency observed on lymphocytes

isolated from healthy donors and cancer patients. In human whole blood, ex

vivo addition of AB154 achieved complete inhibition of TIGIT.

Figure 4. In T cells isolated from advanced head and neck tumor samples (n =

5), markers of antigen experience are predominantly found on the CD8+ subset.

CD8+ antigen-experienced T cells (CD103+CD39+) express higher levels of PD-

1 and TIGIT that are consistent with an “activated” or “exhausted” phenotype.

CD226 is progressively lost from the inexperienced CD103-CD39- population to

the experienced CD103+CD39+ CD8 T cell population. TIGIT receptor density is

often comparable or higher on antigen-experienced CD8 T cells than on Treg.

PD-1 is consistently highest on intra-tumoral CD103+CD39+ CD8 T cells.

Similar AB154 Potency on Peripheral Blood

Lymphocytes Isolated from Healthy Volunteers

and NSCLC Patients

TIGIT and PD-1 Are Highly Expressed by

Intra-tumoral Antigen-Experienced CD8+ T Cells

FSC-A

SS

C-A

FSC-A

FS

C-H

CD56

CD

3

CD56

CD

3

CD4

CD

8

CD4

Fo

xP

3

TregCD8+ T

cells

CD4+

T cellsNK

NKT

T

AB154:

Receptor

Occupancy: 0%

CD3

TIG

IT

100%

AB154α-TIGIT

(saturating)

Pre-dose Low dose

AB154

High dose

TIGIT-

expressing

cell

1) As of the DCO of 16-Sep-2019, 10 participants have been dosed with

AB154 monotherapy (0.5, 1, and 3 mg/kg) and 6 participants with

AB122/AB154 combination therapy (1 and 3 mg/kg).

2) No DLTs have been observed with either regimen. DLT evaluation of the

fully enrolled AB154 10 mg/kg + AB122 combination cohort is ongoing.

3) No subject has experienced any AB154-related SAEs or ≥Gr3 AEs.

Total Receptor Coverage Achieved in AB154

Monotherapy and Combination Cohorts

NSCLC (Adeno)

HNSCC (Squam)

TNBC NSCLC (Squam)

Cervical

0

10

20

30

40

50

60

70

80

90

100

CD

15

5

Ex

pre

ss

ion

(% P

osit

ive P

ixel)

Figure 7. Complete receptor coverage has been observed in all TIGIT-

expressing peripheral lymphocytes after initial dosing at every trough timepoint

for all enrolled subjects. PK data for the second monotherapy cohort (AB154 at

1 mg/kg Q2W) is shown. Spikes in Ki67 expression occur between Day 3 and

Day 29 in CD8+ T cells for both monotherapy and combination subjects. Where

present in this small number of subjects, the magnitude of this proliferative burst

appears to increase in a AB154 dose-dependent manner, which may also be

enhanced by combination with AB122 (anti-PD-1).

Safety and Tolerability of AB154

Demonstrated in Ongoing Phase 1 Study

Enhanced Effector Function Results in NK-

Mediated Killing of Activated TIGIT+ CD8+ T Cells

Strong CD155 Staining in Tumors of Interest

Figure 5. Activated CD8+ T cells expressing high levels of TIGIT are particularly

susceptible to antibody-dependent cell-mediated cytotoxicity (ADCC) when

targeted by an antibody with enhanced effector function (EEF). Data from four

combinations of T cell donors (n = 2) and NK cell donors (n = 2) are shown.

*p ≤ 0.05. **p ≤ 0.01. ***p ≤ 0.001. ****p ≤ 0.0001.

Pre-d

ose

C1D

1 1h

r

C1D

3

C1D

8

C1D

15

C2D

29

C2D

31

C2D

36

C3D

57

C4D

85

0

20

40

60

80

100

AB154 (0.5 mg/kg Q2W)

TIG

IT R

ecepto

r

Occupancy %

Trough Drug Levels

AB154 infusion

Complete Target CoverageAfter First Dose of AB154

AB154

TIGIT

0.5 mg/kg 1 mg/kg 3 mg/kg 1 mg/kg 3 mg/kg1

2

3

4

5

Proliferation in CD8+ T cells

Pe

ak F

old

Ch

an

ge

(Ki6

7+

%,

Da

y 3

- 2

9)

AB154(Monotherapy)

AB154 +AB122 (240 mg)

0 7 14 21 28 35 42

1

10

100AB154 (1 mg/kg Q2W)

Time from First Dose (Days)

AB

15

4 (

g/m

L)

Figure 3. CD155 immunohistochemistry (IHC) shows membrane and

cytoplasmic localization on cancerous cells, blood vessels, and tertiary

lymphoid structures (TLS) in NSCLC and other tumors of interest. Arrows

indicate positive staining. CD155 protein levels were quantified from IHC.

No

Ab

hIgG

1.Fc

Sile

nt

AB15

4

hIgG

1.W

T

AB15

4-W

T

AB15

4-EEF

0

10

20

30

40

% D

ea

d (T

ce

lls)

****

********

ns

ns

**

Activated

CD8+ T cell

NK cell

4 h

37 °CT

IGIT

DNAM-1

Pre-activation Post-activation

Resting

CD8+ T cell

α-CD3/CD28

beads

ADCC

0

20

40

60

80

100

TIG

IT+

%

Healthy

NSCLC

CD8+

T cellsCD8

-

T cellsNKT NKTotal

LymphocytesT cells

0.01 0.1 1 10 100

0

10

20

30

40

AB154-AF647 (nM)

EC50 = 0.13 nM

0.01 0.1 1 10 100

0

10

20

30

40

AB154-AF647 (nM)

TIG

IT+

%

(To

tal L

ym

ph

ocyte

s)

EC50 = 0.14 nM

CD226

TIGIT CD155

AB154

GzmBIFNg

GzmBIFNg

Cytotoxic Granules

CD226

TIGITCD155

SuppressedImmune Cell

ActivatedImmune Cell

Tumor Cell T Cell/NK CellT Cell/NK Cell

* Data collected as of 8-Oct-2019

TGCT: Testicular Germ Cell Tumors; LUAD: Lung Adenocarcinoma; HNSC: Head and Neck Squamous Cell

Carcinoma; CESC: Cervical Squamous Cell Carcinoma; LUSC: Lung Squamous Cell Carcinoma

Tumor

Stroma

vessels

Tumor

TLS

NSCLC

HNSCC CervicalTNBC

NSCLCNSCLC

C1D15 C2D29

Predose Predose

1 0.5 mg/kg 99.7 ± 0.3 100 ± 0

2 1 mg/kg 100 ± 0 100 ± 0.12

3 3 mg/kg 100 ± 0 100 ± 0

1 1 mg/kg 100 ± 0 99.6 ± 0.75

2 3 mg/kg 100 ± 0 100 ± 0

3 10 mg/kg 100 TBD

AB154

Monotherapy

AB154 + AB122

(240 mg)

Combination

Receptor Occupancy

(Mean+SD%)Study Arm Cohort

AB154

Dosing

(Q2W)

TIGIT PD-1 CD226

TIGIT

CD

22

6

CD45

Via

bili

ty

FSC-A

CD

3

CD8

FoxP

3

CD103

CD

39

TregCD4+

CD8+

Treg

CD8+

CD4+

CD103

CD103

CD8+

Treg CD4+

0

50

100

% o

f C

D3

+

TIGIT PD-1 CD226

0

5×103

1×104

1.5×10 4

Ge

oM

ea

n o

f

CD

8+ T

ce

lls

0

20

40

60

80

100

% o

f C

D8

+ T

ce

lls

CD226-

TIGIT+

CD226+

TIGIT+

CD226-

TIGIT-

CD226+

TIGIT-

CD103-CD39

-

CD103+

CD39-

CD103+

CD39+

A B C D E

0

3000

6000

9000

12000

15000

PD

-1 M

FI

CD103+ CD39+

CD8+ T cells

Treg

A B C D E

0

3000

6000

9000

12000

TIG

IT M

FI

Subject Subject

CD103+ CD39+

CD103- CD39-

CD103+ CD39-

CD8+ T Cells