Science - USA (2020-08-21)

sciencemag.org SCIENCE

GRAPHIC: MELISSA THOMAS BAUM/

SCIENCE

product cyclic di-GMP but unlike endoge-
nous cGAMP ( 14 ). The agonists presented
by Pan et al. and Chin et al. both stabi-
lize the closed conformation of STING.
Further study is necessary to tease apart
the biological consequences of stabilizing
STING in its open versus closed confor-
mations. The MSA-2 compound described
by Pan et al. also demonstrated limited
toxicity in mice despite systemic adminis-
tration, owing to its preferential bioactiv-
ity within the acidic milieu of the tumor
microenvironment.
A second important consideration is
the effect of systemic STING agonists on
specific immune cell subpopulations. Chin
et al. noted that SR-717 induced expres-
sion of the immunosuppressive molecules
PD-L1 and indoleamine 2,3-dioxygenase 1
(IDO1) in primary human peripheral blood
mononuclear cells in vitro. Additionally,
intraperitoneal injection of SR-717 in a
melanoma mouse model led to increased
PD-L1 expression on CD11c+CD8– dendritic
cells but not on CD8+ dendritic cells iso-
lated from tumor-draining lymph nodes.
Although CD8+ dendritic cells are thought
to be the key APC subset for inducing tu-
mor-specific CD8+ T cell priming, it is nota-
ble that SR-717 affected these dendritic cell
subtypes differently. Further characteriza-
tion of the ways by which STING agonists
induce both stimulatory and suppressive

events in relevant cell subpopulations

within the tumor microenvironment will

be critical. Antitumor efficacy of SR-717

was not improved by either anti–PD-1 or

anti–PD-L1 treatment in a mouse model of

melanoma, which is in contrast to MSA-2,

which did show improved tumor shrinkage

when combined with anti–PD-1 therapy.

These differences could be due to differ-

ent molecular properties of these STING

agonists, differences in dose and schedule

of administration in combination with im-

mune checkpoint blockade, or distinctions

between the experimental models used.

A third consideration for clinical devel-

opment is the dose and schedule of admin-

istered drug. These need to be optimized

carefully, because systemic administra-

tion also may give rise to a bell-shaped

efficacy curve. Probing pharmacodynamic

endpoints within the tumor microenviron-

ment associated with activity should guide

selection of therapeutic dosing. Fourth,

the consideration of which tumor types

and which patients have the potential to

respond to these agents also needs to be

addressed, so predictive biomarkers for ap-

propriate patient selection also need to be

pursued. A final consideration is that other

innate immune agonists are advancing

in clinical development, including agents

targeting TLR pathways, such as TLR9

( 15 ). Understanding which innate im-

mune pathway is functionally relevant in

distinct patient populations will be para-

mount toward optimization of innate im-

mune agonist combinations with existing

immunotherapies.

The compounds reported by Chin et al.

and Pan et al. illustrate how distinctive mo-

lecular properties of STING agonists can

determine the balance of activity in the

tumor versus systemically. Non-nucleotide

small-molecule STING agonists that can be

administered systemically may represent an

attractive approach for targeting this path-

way and have the potential to transform the

therapeutic landscape once optimized. j

REFERENCES AND NOTES

1. G. J. Freeman et al., J. Exp. Med. 192 , 1027 (2000).


2. M. Ayers et al., J. Clin. Invest. 127 , 2930 (2017).


3. P. C. Tumeh et al., Nature 515 , 568 (2014).


4. B.-S. Pan et al., Science 369 , aba6098 (2020).


5. E. N. Chin et al., Science 369 , 993 (2020).


6. R. K. Vaddepally, P. Kharel, R. Pandey, R. Garje, A. B.
   Chandra, Cancers (Basel) 12 , 738 (2020).


7. J. A. Trujillo, R. F. Sweis, R. Bao, J. J. Luke, Cancer
   Immunol. Res. 6 , 990 (2018).


8. S.-R. Woo et al., Immunity 41 , 830 (2014).


9. L. Sun, J. Wu, F. Du, X. Chen, Z. J. Chen, Science 339 , 786
   (2013).


10. L. Corrales et al., Cell Rep. 11 , 1018 (2015).


11. J. Conlon et al., J. Immunol. 190 , 5216 (2013).


12. F. Meric-Bernstam et al., J. Clin. Oncol. 37 (15_suppl.),
    2507 (2019).


13. K. J. Harrington et al., Ann. Oncol. 29 , viii712 (2018).


14. J. M. Ramanjulu et al., Nature 564 , 439 (2018).


15. M. Reilley et al., J. Clin. Oncol. 37 (15_suppl.), TPS2669
    (2019).

ACKNOWLEDGMENTS

The authors are funded by the National Institutes of Health

(grants F30CA250255 to E.F.H. and R35CA210098 to T.F.G.).

T.F.G. reports a licensing agreement and receives research

support and consultancy fees from Aduro Biotech.

10.1126/science.abc6622

INSIGHTS | PERSPECTIVES

F

F

O

O

N

NN

N N

F

O

O

F

F FO

O

N N

N

NN

NNN

O

O

O

O

O S O

O

O

O

S

SS

N

NN

N

O

O

O O

O

O O

F

CD80

STING

Open

conformation

STING

Closed

conformation

CD86

cDC1 dendritic cell activation CD8+ T cell cross-priming

Tumor shrinkage

IFN-b

I L- 6

MSA-2

SR-717

Department of Pathology, University of Chicago, Chicago,

IL, USA. Email: [email protected]

New innate immune agonists
The non-nucleotide stimulator of interferon genes (STING) agonists MSA-2 and SR-717 reported by Pan et al. and Chin et al., respectively, stabilize STING in its closed
conformation. STING activation induces downstream signaling events that culminate in the expression of inflammatory cytokines such as interferon-b (IFN-b) and
interleukin-6 (IL-6). Secretion of these cytokines in the tumor microenvironment promotes the maturation and activation of cDC1 dendritic cells, which then promote
antitumor immunity by priming tumor antigen–specific CD8+ T cells in the tumor-draining lymph node.

922 21 AUGUST 2020 • VOL 369 ISSUE 6506

Published by AAAS