RESEARCH POSTER PRESENTATION DESIGN © 2015

www.PosterPresentations.com

• Robust assays have been developed to measure the peripheral levels and activity of the

adenosine-generating enzyme, CD73.

• Immunohistochemical methods have been developed to measure and quantify adenosine-

generating enzymes CD73 and TNAP in human tumor tissue.

• These assays are being implemented to assess the adenosine generating potential of subjects

in clinical studies with AB928.

• AB928 is currently in clinical trials in combination with standard of care chemotherapeutics and

an anti-PD-1 immunotherapeutic.

DiRenzo D, Ashok D, Anderson AE, Udyavar A, Park A, Tan JBL, Luu I, Zhang K, Jeffrey JL, Seitz L, Leleti MR, Young SW, Powers JP, Walters MJ

Arcus Biosciences, Inc.; 3928 Point Eden Way, Hayward, CA 94545 (USA)

Methods for Assessment of the “Adenosine Fingerprint”

in Clinical Trials of AB928

pCREB Assay: Whole blood from healthy volunteers or cancer subjects was stimulated ex vivo

with the adenosine agonist NECA. Levels of CREB phosphorylation (pCREB) were assessed by

flow cytometry and AB928 plasma concentrations were determined using LC-MS/MS.

CD73 ELISA: Circulating levels of CD73 were quantified in matched serum and plasma samples

from healthy donors and cancer patient samples with an in-house ELISA. Validation studies were

performed to correlate CD73 levels between serum and plasma and to ensure acceptable assay

parallelism and inter-assay variation.

AMPase Activity: AMP hydrolytic activity in serum from healthy volunteers or patients with solid

tissue tumors was assessed using the AMP GloTM (Promega) assay. CD73 and TNAP inhibitor

cocktails were used to assess the relative contribution of each enzyme to AMPase activity.

Gene Expression: Expression of CD73 (NT5E) and TNAP (ALPL) on select tumor types were

derived from RNASeq in The Cancer Genome Atlas (TCGA) database. Data is displayed as log2

transformed expression of counts per million.

Histology: Immunohistochemistry (IHC) was performed using CD73 (Cell Signaling, D7F9A)

and TNAP (Sino Biological, R034) on formalin fixed paraffin embedded (FFPE) human tissue.

QuPath software was used for quantification and H-score = 3x(high intensity staining area) +

2x(medium intensity staining area) + (low intensity staining area). .

Introduction

Figure 1. (A.) AB928 suppresses NECA-induced pCREB activation in CD8+ T cells of healthy

volunteers. (B.) Healthy volunteers and cancer patients exhibit a similar pattern of pCREB

inhibition and an overlapping PK/PD relationship.

Results

Materials and Methods

Conclusions

AACR 2019 National Meeting; Atlanta

Abstract # 3168

The tumor microenvironment (TME) contains high levels of immunosuppressive adenosine

(ADO), which activates the A2aR and A2bR receptors on immune cells, leading to an ineffective

anti-tumor response. Ecto-5’-nucleotidase (CD73) and tissue non-specific alkaline phosphatase

(TNAP) are primarily responsible for the conversion of extracellular adenosine mono-phosphate

(AMP) to ADO and exhibit both membrane-bound

and secreted forms. We have previously shown

that AB928, a dual A2aR/A2bR antagonist,

rescues the immuno-suppressive effects of ADO

in experimental tumor models. Herein, we

describe the development of assays to measure

the expression and activity of adenosine-

generating enzymes in human tumor samples

and peripheral blood. These assays are being

used to define an “adenosine fingerprint” to

identify tumor types and patients most sensitive

to adenosine inhibition by AB928.

0

50

100

% pC

RE

B S

ignal

Pre-dose Day 4Pre-dose

Day 42 hours

Day 424 hours

Placebo 10 mg

75 mg 150 mg

25 mg

200 mg

1000 2000 3000 4000

0

50

100

AB928 measured in plasma (nM)

%

Inhib

itio

n fro

m P

redose

Cancer patientsHealthy volunteersA. B.

Development of a Validated ELISA Assay to Quantify Soluble

CD73 in Peripheral Blood

Figure 3. (A.) Healthy volunteer serum was tested in the presence of CD73 and/or TNAP

inhibitors. (B.) Average CVs across multiple conditions show the assay is robust for the

determination of AMP hydrolysis in human serum. (C.) Healthy volunteer and cancer patient

serum display a strong correlation between CD73 protein concentration and AMP hydrolysis.

Determination of AMP Hydrolysis by CD73 and TNAP in Serum

Using an AMP GloTM Assay

Figure 2. (A.) Serum and sodium heparin plasma from healthy donors were evaluated in an

ELISA to measure soluble CD73 and were highly correlated. (B.) Parallelism assessment was

performed to identify the quantitative range of the assay. (C.) Soluble CD73 levels in serum are

elevated in cancer patients compared to healthy controls. *p<0.05

1 2 3 4 51

2

3

4

5

Plasma vs. Serum

Serum CD73 [ng/ml]

Pla

sm

a C

D73 [ng/m

l]

R2 = 0.9597

0 2 4 6 8 10 12 14 16 18 2040

60

80

100

120

140

160

180

Parallelism Assessment

Sample Dilution Factor

Adju

ste

d %

Recovery

Below Detection

2

4

5

8

10

MRD

Healthy Cancer0

2

4

6

8

10

Seru

m C

D73 [ng/m

l]

*A. B. C.

TCGA Analysis Reveals Unique Expression Patterns of CD73 and

TNAP in Human Tumors

Figure 4. CD73 (left) and TNAP (right) expression from RNAseq data retrieved from The

Cancer Genome Atlas (TCGA) samples. Numbers indicate a ratio of log2 counts per million per

sample. Samples are ordered according to their expression with higher expressing tumors on

the left and lower expressing tumors on the right.

TNAP Expression is Heterogeneous Across Human Tumors from

Different Indications

Figure 5. (A.) Representative images of immunostaining for CD73 (brown) on human FFPE

tumor samples. Nuclei are stained with hematoxylin (blue) (B.) Quantification of CD73 staining

area as a percentage of total tumor area. (C.) A strong correlation is seen when comparing an

H-score to % staining area.

CD73 is Detected and Quantifiable in Human Tumors via IHC

NSCLC NSCLC TNBC CRC

NSCLCadeno

CRC HNSCC NSCLCsquam

Gastric Esoph Breast0

10

20

30

40

% C

D73

+ a

rea

0 20 40 60 80 100

0

20

40

60

80

100

% CD73+ Area

H s

co

re

A.

B. C.

Ovarian NSCLC(adeno)

Breast Liver Stomach CRC0

25

50

75

100

% T

NA

P+ A

rea

Figure 6. (A.) Representative images of immunostaining for TNAP (brown) on human FFPE

tumor samples. Strong staining can be seen in ovarian and NSCLC tumors whereas liver and

CRC exhibit much less staining. (C.) Quantification of TNAP staining area as a percentage of

total tumor area.

A.

C.

AB928 Inhibits A2aR Activation Following Dosing with ≥75 mg in

Healthy Volunteers and Cancer Subjects

0

3×105

6×105

9×105

1.2×106

AMP Hydrolysis in Healthy Serum

Lum

inescence (

RLU

)

CD73 Inhibitor

TNAP Inhibitor +

+ +

+

--

- -0 2 4 6 8

0

20

40

60

80

100

CD73 Activity vs. Protein Correlation

CD73 [ng/ml]

% H

ydro

lytic A

ctivity

R2 = 0.7151

TNBC

R2 = 0.9532

Healthy

NSCLC

CRC

A.

B. C.

Ovarian NSCLC Breast Liver

CRCStomachB.

Interested in a career at Arcus? Visit www.arcusbio.com

CD73 TNAP

Teff