Tumor cell intrinsic STING signaling demonstrates minimal contribution to the anti-tumor response elicited by the STING agonist ADU-S100 (MIW815)Leticia Corrales1, Anthony L. Desbien1, Weiwen Deng1, Kelsey Sivick Gauthier1, Tamara Schroeder1, Gabrielle Reiner1, Natalie H. Surh1, Brian Francica1, Ken Metchette1, Chudi O Ndubaku1, Jeffrey M McKenna², Yan Feng², Lianxing Zheng², Steven L Bender³, Charles Y Cho³, Meredith Leong1, Andrea van Elsas1, and Sarah M. McWhirter1

1 Aduro Biotech, Inc., Berkeley, CA; 2 Novartis Institutes for BioMedical Research, Cambridge MA; 3 Genomics Institute of the Novartis Research Foundation, San Diego, CA

Poster# 1202

Abstract

Innate immune sensing of tumors is a critical step in generating spontaneous anti-tumor T cell responses. Endogenous activation of the STING pathway in immune cells and the subsequent generation of type I IFN is sufficient to generate spontaneous anti-tumor T cell responses. To take advantage of this tumor defense mechanism for therapeutic intervention, Aduro has developed the first-in-class STING agonist, ADU-S100 (MIW815), a small molecule derivative of the natural cyclic dinucleotide STING ligand. In mouse models, intratumoral administration of ADU-S100 (MIW815) increases systemic tumor-specific T cells and results in substantial antitumor efficacy1-2. ADU-S100 (MIW815) is currently being tested as mono- and combination therapy in Phase 1 clinical studies enrolling patients with cutaneously-accessible treatment- refractory advanced solid tumors and lymphomas. STING is broadly expressed across different cell types, and several studies have demonstrated that tumor cells regulate expression of STING and other members of the pathway, mainly cGAS, by epigenetic mechanisms3-6. Activation of the STING pathway in innate immune cells is necessary for the generation of anti-tumor T cell responses, but the role of other cell types within the tumor microenvironment in response or resistance to STING agonists is not completely understood. To understand the contribution of tumor-cell STING to the anti-tumor response, we generated STING-deficient 4T1 tumor cells. Composition of the tumor microenvironment, endogenous T cell responses and tumor growth were comparable in animals with implanted STING-WT or -KO tumor cells. Similar data was observed in the B16.SIY melanoma model. In order to understand the role of tumor-STING in the context of ADU-S100 (MIW815) treatment, animals implanted with 4T1 STING WT or KO tumor cells were treated with a wide range of ADU-S100 (MIW815) doses. Expression of STING within tumor cells did not impact activation of innate cells or generation of tumor-specific T cells across all the tested doses. At immunogenic doses, tumor cell expression of STING was not required for tumor growth control.Overall, these results show that ADU-S100 (MIW815) activation of STING in host cells rather than in tumor cells is critical for producion of type I interferon and tumor control. This supports potential treatment of cancers of different histologies regardless of tumor cell-intrinsic STING expression.

Summary

STING deficient 4T1 and B16.SIY tumor cell lines were generated. Pools of 4T1 STING-/- clones failed to generate cytokines or overexpress upregulate expression of MHC I in vitro after stimulation with ADU-S100 (MIW815).

Injection of STING-sufficient or deficient tumor cell lines in vivo generated tumors with a comparable composition of innate cells and specific CD8+ T cells, induced similar systemic anti-tumor CD8 responses and presented similar growth rates.

STING expression in tumors is not required for the generation of systemic T cell responses and tumor control after IT injection of immunogenic doses of ADU-S100 (MIW815).

References

1 Corrales, L. et al. Direct Activation of STING in the Tumor Microenvironment Leads to Potent and Systemic Tumor Regression and Immunity. Cell Rep 11, 1018-1030, (2015).

2 Sivick, K. E. et al. Magnitude of Therapeutic STING Activation Determines CD8(+) T Cell-Mediated Anti-tumor Immunity. Cell Rep 25, 3074-3085 e3075, (2018).

3 Xia, T., Konno, H., Ahn, J. & Barber, G. N. Deregulation of STING Signaling in Colorectal Carcinoma Constrains DNA Damage Responses and Correlates With Tumorigenesis. Cell Rep 14, 282-297, (2016).

4 Xia, T., Konno, H. & Barber, G. N. Recurrent Loss of STING Signaling in Melanoma Correlates with Susceptibility to Viral Oncolysis. Cancer Res 76, 6747-6759, (2016).

5 Kitajima, S. et al. Suppression of STING Associated with LKB1 Loss in KRAS-Driven Lung Cancer. Cancer Discov 9, 34-45, (2019).

6 Konno, H. et al. Suppression of STING signaling through epigenetic silencing and missense mutation impedes DNA damage mediated cytokine production. Oncogene 37, 2037-2051, (2018).

Lack of STING in tumor cells impairs the production of cytokines in the context of ADU-S100 in vitro

IFN

β (p

g/m

l)

S100 (µg)

50

40

30

20

10

0

4T1 STING+/+

4T1 STING-/-

S100 (uM)

0.00

0.08

0.46

2.78

16.67

100.00

MFI: 4000 10000

4T1 STING+/+ 4T1 STING-/-

GMCSFIFNγ

IL-1βIL-12p70

IL-13IL-18IL-2IL-4IL-5IL-6

TNFαENA78GCSFIFNαIL-1α

IL-15/RIL-28IL-3

IL-31LIF

MCSFIL-10

IL-17αIL-22IL-23IL-27IL-9

EOTAXINGRO-ALPHA

IP10MCP1MCP3MIP1aMIP1βMIP2

RANTES

in vitroADU-S100

0 100

4T1 STING+/+ 4T1 STING-/-

S100 (µM ) 0 10 100 0 10 100

4T1 STING+/+ 4T1 STING-/-

pSTING

total STING

pTBK1

total TBK1

pIRF3

total IRF3

ATPase

500

400

300

200

100

0B16.SIY MC38 CT26 4T-1 Gt BMM

IFN

-β m

RN

A (fo

ld to

non

-trea

ted)

0 µM S10050 µM S100

6 9 16 20 23 26

1500

1000

500

0

Tum

or v

olum

e (m

m3 )

Day post tumor implantation13 29 33

B16.SIY STING+/+

B16.SIY STING-/-

ns

8 11 15 18 22 25

1500

1000

500

0

Tum

or v

olum

e (m

m3 )

Day post tumor implantation

ns

4T1 STING+/+

4T1 STING-/-

1.5

1.0

0.5

0.0AH1+ (

% o

f CD

8+ cel

ls)

Spleens

ns15

10

05

0

Tumors

ns

AH1+ (

% o

f CD

8+ cel

ls) 20

15

10

5

0Tota

l CD

8+ / tu

mor

vol

ume ns

4T1 STING+/+ 4T1 STING-/- 4T1 STING+/+ 4T1 STING-/- 4T1 STING+/+ 4T1 STING-/-

MacrophagesNeutrophils

Monocytes

NK cells

100

80

60

40

20

0

% o

f CD

45+ c

ells

4T1 STING+/+ 4T1 STING-/-

Tumor-cell expression of STING is dispensable for the endogenous generation of a tumor-specific T cell

response and tumor control in vivo

L

GMSCFIFNγIL1β

IL12p70IL-13IL-18IL-2IL-4IL-5IL-6

TNFαENA78GCSFIFNαIL-1α

IL-15RIL-28IL-3

IL-31LIF

MCSFIL-10

IL-17αIL-22IL-23IL-27IL-9

EOTAXINGRO-ALPHA

IP10MCP1MCP3MIP1αMIP1βMIP2

RANTES

PBS (5h) PBS (24h) S100 (24h)S100 (5h)STING+/+ STING-/- STING+/+ STING-/- STING+/+ STING-/- STING+/+ STING-/-

0 100

STING expression in tumor cells is dispensable for the changes in the TME generated by ADU-S100

S100 - + - + - + - + - + - +

NeutrophilsMonocytes Macrophages

% o

f CD

45+ c

ells

30

20

10

0

4T1 STING+/+

4T1 STING-/-

30

20

10

0

40

80

60

20

0

100

40

ns

******

ns

ns

ns

*** **

****ns

ns

7 10 15 18 22 24Day post tumor implantation

28

1500

1000

500

0

Tum

or v

olum

e (m

m3 ) 2000

PBS

ns

Day post tumor implantation

1500

1000

500

0

Tum

or v

olum

e (m

m3 ) 2000

1 µg S100

ns

7 10 15 18 22 24 28

B16.SIY STING+/+

B16.SIY STING-/-

8 11 15 18 22 25

1500

1000

500

0Tum

or v

olum

e (m

m3 )

Day post tumor implantation29

2000

ns

PBS

8 11 15 18 22 25

1500

1000

500

0Tum

or v

olum

e (m

m3 )

Day post tumor implantation29

2000

ns

10 µg S100 4T1 STING+/+

4T1 STING-/-

0 1 100 500S100 (µg)

4

2

1

0AH

1+ (%

CD

8+ cel

ls)

3

10

ns

ns ns

nsns

0 0.1 10 100S100 (µg)

15

10

5

SIY+ (

% C

D8+ c

ells

)

01

ns

ns

ns

ns ns

4T1 STING+/+

4T1 STING-/-

B16.SIY STING+/+

B16.SIY STING-/-

STING expression in tumor cells does not affect anti-tumor specific T cell responses or control of tumor growth at immunogenic doses of ADU-S100

Figure 1. B16.SIY, MC38, CT26, 4T1 tumor cells and STING-deficient bone marrow macrophages (BMM) derived from Goldenticket mice were stimulated with 50 µM ADU-S100. Total RNA was purified, retrotranscribed and the expression of IFN-β was quantifiedy by real-time PCR (A). 4T1 STING+/+ and STING-/- cells were generated and pools of clones of 4T1 STING+/+ and STING-/- were stimulated with 10 or 100 µM ADU-S100. The amount of pSTING, pTBK1, pIRF3, total STING, total TBK1 and total IRF3 proteins was examined by western blot (B). Pools of clones of 4T1 STING+/+ and STING-/- were stimulated with a wide range of ADU-S100 concentrations. After 24 hours cell supernatants were collected, measurement of cytokines and chemokines were assessed by Luminex (C), INF-β generation was measured by ELISA (D), and expression of MHC I on cell surface was analyzed by flow cytometry (E).

Figure 3. WT Balb/c mice bearing a single flank tumor of 4T1 STING+/+ or 4T1 STING-/- tumor cells received one IT injection of vehicle alone or 10 µg ADU-S100. 5 and 24 hours post-injection tumors were collected, and measurement of cytokines and chemokines in the supernatant was performed by Luminex (A). 24 hours post-treatment percentage of CD11b+Ly6C+MHC Class II+ monocytes, and CD11b+CD11c+Ly6C-MHC Class II+ macrophages and CD11b+Ly6G+ neutrophils was assessed by flow cytometry (B).

Figure 4. WT Balb/c bearing a single flank tumor of 4T1 STING+/+ or 4T1 STING-/- tumor cells received one IT injection of vehicle alone or 1, 10, 100, or 500 µg ADU-S100. 8 days post-treatment PBMCs from each mouse were assessed for the frequency of CD8+ H-2Ld-AH1 tetramer+ (AH1+) cells by flow cytometry (A). WT C57BL/6 mice bearing a single flank tumor of B16.SIY STING+/+ or B16.SIY STING-/- tumor cells received one IT injection of vehicle alone or 0.1, 1, 10 or 100 µg ADU-S100. 8 days post-treatment PBMCs from each mouse were assessed for the frequency of CD8+ H-2Kb-SIY tetramer+ (SIY+) cells by flow cytometry (B). Tumor volume was measured at the indicated time points (C, D).

Figure 2. WT Balb/c mice were s.c. implanted with 4T1 STING+/+ or 4T1 STING-/- tumor cells in the right flank. 8 days post-implantation the frequencies of NK1.1+ NK cells, CD11b+Ly6G+ neutrophils, CD11b+Ly6C+MHC Class II+ monocytes, and CD11b+CD11c+Ly6C-MHC Class II+ macrophages in tumors (A); and frequency of CD8+ H-2Ld-AH1 tetramer+ (AH1+) cells in tumors and spleens (B), were assessed by flow cytometry. 4T1 STING+/+ or STING-/- tumor cells or B16.SIY STING+/+ or STING-/- tumor cells were implanted in the right flank of Balb/c mice and tumor volume was measured at the indicated time points (C).

C

BACBA

D E

BA CA

B D