Guide: Mouse models for the assessment of cGAS/STING therapeutics: practical considerations for translational research

Messages clés (résumé)

• The cGAS-STING pathway detects misplaced double-stranded DNA and connects DNA sensing to type I interferon and inflammatory signaling.
• cGAS and STING are distinct pharmacological targets:
  
  • cGAS inhibitors prevent the production of 2′3′-cGAMP.
  • STING inhibitors block signaling downstream of cGAMP.
  • STING agonists activate innate immune signaling independently of upstream DNA detection.
• Pathway activation can support antitumor immunity, antiviral responses and vaccine adjuvanticity.
• Excessive or chronic activation can contribute to autoinflammatory, autoimmune and neuroinflammatory disease.
• Wild-type mice are valuable for pathway biology but may not predict the activity of compounds designed specifically for human cGAS or human STING.
• STING pharmacology is particularly species- and allele-dependent. Activity should be confirmed against the exact human STING variant being modelled.

‍

‍

Introduction / Contexte

What is the cGAS-STING pathway and how does it work?

The cGAS-STING pathway is an innate immune surveillance system that detects DNA in cellular locations where DNA should not normally be present.

cGAS is the DNA sensor and cGAMP-producing enzyme. STING is the downstream adaptor and signaling hub.

‍

How does cGAS detect cytosolic DNA and activate STING?

• cGAS binds directly to double-stranded DNA. DNA binding promotes the formation of an active cGAS-DNA complex.
• Activated cGAS catalyzes 2′3′-cGAMP production from ATP and GTP, which diffuses within the cell and may also be transferred between cells. cGAMP binding activates STING and initiates its intracellular trafficking.
• Activated STING recruits TBK1, leading to IRF3 activation and transcription of type I interferon-associated genes.

‍

How is the pathway activated and regulated?

Activating inputs

• Microbial DNA and mitochondrial DNA release
• Cytosolic chromatin fragments
• Micronuclei generated by genomic instability
• Radiation or DNA-damaging treatment
• Cellular senescence
• Direct STING agonists

Regulatory mechanisms

The duration and amplitude of signaling are influenced by:

• Removal or degradation of cytosolic DNA
• cGAMP degradation
• STING trafficking
• STING post-translational modification
• Autophagic and lysosomal processes
• Feedback through interferon and inflammatory pathways
• Cell type, activation state and tissue context

Why regulation matters

• Acute, local activation may support immunity and tumor rejection.
• Persistent or systemic activation may produce inflammation, tissue injury or counter-regulatory immune suppression.
• Chronic STING activation can engage outputs that differ from a brief canonical type I interferon response (Vasiyani, 2026; Szego et al., 2022).

‍

Which downstream immune responses are triggered?

Potential responses include:

• Type I interferon production
• Interferon-stimulated gene expression
• CXCL10, CCL5 and other chemokines
• Dendritic-cell activation and antigen presentation
• Recruitment and activation of monocytes and other myeloid cells (Ren et al., 2025)
• Natural killer cell activation
• CD8-positive T-cell priming and infiltration
• Changes in tumor-associated macrophage states
• NF-κB-dependent inflammatory cytokines
• Autophagic, inflammasome or cell-death responses in some contexts

In tumor models, STING activation can recruit inflammatory monocyte-lineage cells, alter macrophage programs and support CD8-positive T-cell and natural killer cell activity. The dominant response nevertheless depends on agonist chemistry, cellular target, dose, route and tumor model (Girard et al., 2025; Appleman et al., 2025).

‍

‍

Why therapeutically target the cGAS-STING pathway?

What is the pathway’s role in cancer and antitumor immunity?

Potential antitumor effects include:

• Sensing tumor-derived DNA.
• Inducing type I interferons (Banerjee et al., 2026)
• Activating dendritic cells.
• Supporting tumor-antigen cross-presentation.
• Priming tumor-specific CD8-positive T cells.
• Recruiting effector immune cells.
• Converting an immune-poor tumor into a more inflamed microenvironment.

However, the pathway is context-dependent:

• Chronic signaling may promote tissue inflammation without effective tumor rejection.
• Tumors may suppress or lose pathway components.
• Strong systemic activation can limit the therapeutic window.
• Some signaling contexts can support immune escape, stromal remodeling or metastasis.

The pathway should therefore be treated as a context-dependent regulatory axis, not as a universal immune “on switch” (Wu et al., 2026; Wang et al., 2025).

‍

What is its role in inflammation and autoimmune disease?

Abnormal self-DNA accumulation can chronically activate cGAS-STING.

Potential disease-relevant triggers include:

• Defective nucleic-acid clearance
• Mitochondrial damage
• Genomic instability
• Cellular senescence
• Constitutively active STING variants
• Persistent tissue injury

This provides a rationale for cGAS or STING inhibition in selected autoinflammatory, autoimmune and neuroinflammatory settings. Chronic activation must be evaluated separately from acute pharmacological activation because the affected cells, duration and downstream outputs may differ (Seok et al., 2023; Vasiyani, 2026; Szego et al., 2022).

Diseases linked to the cGAS-STING pathway include STING-associated vasculopathy with onset in infancy (SAVI) and systemic lupus erythematosus (SLE).

‍

Can STING agonists enhance checkpoint-inhibitor responses?

Yes, preclinical studies show that STING agonism can complement PD-1 or PD-L1 pathway blockade by:

• Increasing inflammatory chemokines.
• Improving effector-cell recruitment.
• Supporting antigen presentation.
• Increasing T-cell priming.
• Reprogramming myeloid populations.

Systemically delivered, tumor-targeted STING agonist conjugates have produced combination activity with anti-PD-L1 in syngeneic mouse models, while reviews of clinical development indicate that durable patient benefit remains more difficult to achieve than in preclinical models (Wu et al., 2022; Shi et al., 2025).

‍

How does STING activation modify the tumor microenvironment?

Possible changes include:

• Increased type I interferon signaling, as shown using a human STING agonist in the genO-hSTING model (Banerjee et al., 2026)
• Increased CXCL10 and other immune-recruiting chemokines.
• Activation of dendritic and myeloid cells (Ren et al., 2025)
• Increased CD8-positive T-cell and natural killer cell infiltration.
• Reduction or reprogramming of immunosuppressive macrophage states.
• Improved tumor-antigen transport to draining lymph nodes.
• Increased PD-L1 or other adaptive resistance pathways, supporting a rationale for combination therapy.

In preclinical glioblastoma studies, STING agonists were associated with increased CD8-positive T-cell and natural killer cell infiltration, pro-inflammatory macrophage changes and prolonged survival in several models, although study designs and delivery approaches were heterogeneous (Sampath et al., 2026).

‍

Comment choisir le modèle adapté à une étude

Posez-vous d'abord ces questions :

• Is the compound specific for human cGAS, human STING or both?
• Is the intervention an agonist, inhibitor, degrader, antibody or genetic therapy?
• Does the molecule interact with the orthologous mouse target?
• Is the objective pathway biology, pharmacology, efficacy, safety or biomarker qualification?
• Is target activity required in tumor cells, immune cells, stroma, CNS cells or multiple compartments?
• Is the intended treatment acute or chronic?
• Does human STING allele status affect activity?
• Is an intact immune system required?
• Must the model support syngeneic tumors or human tumors?

‍

Comparison of mouse models for cGAS-STING targeted therapies

‍

Exemples concrets

When should a humanized model be used?

Humanized cGAS

Use when:

• The compound directly binds human cGAS.
• Human cGAS target engagement must be measured in vivo.
• Testing a CNS-penetrant human cGAS inhibitor, provided CNS expression and function are validated.

Humanized STING

Use when:

• The agonist or inhibitor is human-STING-specific.
• Human STING alleles influence compound potency.
• A mouse surrogate cannot reproduce the clinical molecule’s pharmacology.
• Human STING activity is required in an intact immune environment.

Double-humanized cGAS/STING

Use when:

• Target engagement depends on human cGAS and human STING.
• Human cGAMP production and downstream human STING pharmacology both matter.
• Pathway-level translation is more important than a single target.
• Combination or indirect mechanisms may engage both proteins.

‍

‍

FAQ

Human versus mouse cGAS-STING biology - what are the principal differences?

Relevant differences may include:

• Protein sequence and structure.
• Ligand-binding pockets.
• Responses to synthetic STING agonists.
• Basal expression across tissues and immune-cell populations.
• Strength and composition of downstream cytokine responses.
• STING trafficking and regulation.
• Human STING allele variation.

‍

Which STING variant should be used: H232 or R232?

Use the variant that best matches:

1. The intended patient population.
2. The compound’s human in vitro pharmacology.
3. The allele used during discovery and candidate optimization.
4. The specific translational question.

Important point

H232 versus R232 should not be selected merely by lab convention. Human STING genetic diversity can affect signaling and may influence translational interpretation. Precision approaches that incorporate human STING variants are increasingly recognized as important for clinical development (Liu et al., 2026; Shi et al., 2025).

‍

Do humanized STING mice reproduce expected cytokine responses?

They can support human-STING-dependent responses if the inserted receptor is:

• Expressed at appropriate levels.
• Present in relevant cell populations.
• Capable of normal trafficking.
• Coupled to the murine downstream signaling machinery.
• Responsive to the selected human agonist.

A cytokine increase alone is insufficient. The response should be linked to exposure, proximal STING engagement and the expected human variant. The genO-hSTING model was demonstrated to reproduce cytokine responses.

‍

Which other therapeutic areas can be studied with humanized cGAS/STING models beyond immuno-oncology?

Humanized cGAS/STING models may support research in:

• Inflammation and autoimmunity
  
  • Test cGAS or STING inhibitors.
  • Measure type I interferon signatures.
  • Connect target engagement with tissue pathology.
  • Evaluate chronic dosing and infection-related safety considerations.

• Vaccine research
  
  • Assess STING agonists as adjuvants.
  • Measure innate activation, antigen-specific antibodies and T-cell responses.
  • Match route and formulation to the intended vaccine.
  • Evaluate local reactogenicity and systemic cytokines.
• CNS and neuroinflammation
  
  • Test brain-penetrant cGAS or STING inhibitors.
  • Measure brain exposure and cell-specific target engagement.
  • Assess microgliosis, astrogliosis and neuronal outcomes.
  • Avoid assuming peripheral biomarkers reflect CNS activity.

Chronic STING activation has been shown to precede dopaminergic neurodegeneration in a constitutively active mouse model, and recent research has implicated cGAS-STING activity in microglial and broader CNS inflammatory responses (Liu et al., 2026; Szego et al., 2022).

• Infectious disease and innate immunity
  
  • Investigate host defense mechanisms.
  • Evaluate the consequences of pathway activation or inhibition.
  • Consider that chronic pathway inhibition may affect antimicrobial immunity.

‍

Does a humanized STING mouse have a human immune system?

Not necessarily. A target-humanized mouse normally has a selected human gene or protein sequence within an otherwise largely murine physiological and immune environment. It should not be described as a human immune system model unless broader immune-system humanization has been performed. However, mice with a human immune system such as the genO-BRGSF-HIS model display functional human STING (Martin et al., 2025; Ren et al., 2025).

‍

Can STING agonists be administered systemically?

Yes, if pharmacology and tolerability permit, but systemic administration requires careful monitoring of:

• Cytokines
• Hématologie
• Liver and other organ toxicity
• Bone-marrow effects
• Tissue distribution
• Maximum tolerated and biologically active doses

Intratumoral delivery offers local exposure but is limited by tumor accessibility. Targeted delivery and non-nucleotide systemic agonists are being investigated to address these limitations (Pan et al., 2020; Huang et al., 2023; Wu et al., 2022).

‍

Why have STING agonists been more successful preclinically than clinically?

Contributing factors include:

• Differences between mouse and human STING pharmacology.
• Dependence on intratumoral administration.
• Limited access to metastatic lesions.
• Narrow systemic therapeutic windows.
• Tumor heterogeneity.
• Immunosuppressive tumor environments.
• Chronic STING activity or adaptive resistance.
• Inadequate patient and biomarker selection.

These gaps reinforce the need for humanized-target models, clinically relevant delivery and pharmacodynamic biomarkers rather than reliance on tumor-volume reduction alone (Xu et al., 2023; Shi et al., 2025).

‍

‍

Références

Appleman VA, Matsuda A, Ganno ML, et al. Selective STING Activation in Intratumoral Myeloid Cells via CCR2-Directed Antibody-Drug Conjugate TAK-500. Cancer Immunol Res. 2025;13(5):661-679. doi:10.1158/2326-6066.CIR-24-0103

Banerjee M, Middya SK, Shrivastava R, et al. C92, a proton channel-blocking allosteric STING agonist generates robust antitumor activity. J Immunother Cancer. 2026;14(6):e015242. Published 2026 Jun 23. doi:10.1136/jitc-2026-015242

Girard M, Yu T, Batista NV, et al. STING agonists drive recruitment and intrinsic type I interferon responses in monocytic lineage cells for optimal anti-tumor immunity. J Immunol. 2025;214(12):3634-3646. doi:10.1093/jimmun/vkaf131

Huang C, Shao N, Huang Y, et al. Overcoming challenges in the delivery of STING agonists for cancer immunotherapy: A comprehensive review of strategies and future perspectives. Mater Today Bio. 2023;23:100839. Published 2023 Oct 21. doi:10.1016/j.mtbio.2023.100839

Liu H, Hu C, Liu H, et al. Mechanistic insights and therapeutic potential of targeting the cGAS-STING pathway in neurodegenerative diseases. J Neuroinflammation. 2026;23(1):221. Published 2026 May 7. doi:10.1186/s12974-026-03815-1

Martin GH, Hedir S, Creusat F, et al. Tumor-dependent myeloid and lymphoid cell recruitment in genO-BRGSF-HIS mice: a novel tool for evaluating immunotherapies. Front Immunol. 2025;16:1624724. Published 2025 Sep 17. doi:10.3389/fimmu.2025.1624724

Pan BS, Perera SA, Piesvaux JA, et al. An orally available non-nucleotide STING agonist with antitumor activity. Science. 2020;369(6506):eaba6098. doi:10.1126/science.aba6098

Ren M, Ai ZE, Zhang Y, et al. cGAS-STING signaling in the tumor microenvironment induces myeloid cell activation and favors T cell-mediated antitumor immunity. Cancer Biol Ther. 2025;26(1):2585562. doi:10.1080/15384047.2025.2585562

Sampath SG, Tang AJ, Chen AX, et al. cGAS-STING agonists in preclinical glioblastoma animal models: a systematic review of tumor microenvironment modulation and survival outcomes. J Neurooncol. 2026;177(3):142. Published 2026 May 8. doi:10.1007/s11060-026-05601-8

Seok JK, Kim M, Kang HC, Cho YY, Lee HS, Lee JY. Beyond DNA sensing: expanding the role of cGAS/STING in immunity and diseases. Arch Pharm Res. 2023;46(6):500-534. doi:10.1007/s12272-023-01452-3

Shi J, Zhang Y, Zhao N, et al. Precision targeting of STING: Challenges, innovations, and clinical outlook for cancer therapy. Innovation (Camb). 2025;7(1):101074. Published 2025 Aug 6. doi:10.1016/j.xinn.2025.101074

Szego EM, Malz L, Bernhardt N, Rösen-Wolff A, Falkenburger BH, Luksch H. Constitutively active STING causes neuroinflammation and degeneration of dopaminergic neurons in mice. Elife. 2022;11:e81943. Published 2022 Oct 31. doi:10.7554/eLife.81943

Vasiyani H. Chronic and non-canonical cGAS-STING activation: implications for health, disease, cancer, and emerging therapeutic opportunities. Apoptosis. 2026;31(2):68. Published 2026 Feb 9. doi:10.1007/s10495-026-02283-5

Wang Q, Yu Y, Zhuang J, Liu R, Sun C. Demystifying the cGAS-STING pathway: precision regulation in the tumor immune microenvironment. Mol Cancer. 2025;24(1):178. Published 2025 Jun 12. doi:10.1186/s12943-025-02380-0

Wu YT, Fang Y, Wei Q, et al. Tumor-targeted delivery of a STING agonist improvescancer immunotherapy. Proc Natl Acad Sci U S A. 2022;119(49):e2214278119. doi:10.1073/pnas.2214278119

Wu S, Tian B, Sun Y, Sun C, Sui B. The cGAS-STING pathway in the tumor immune microenvironment: Multidimensional regulation and therapeutic implications. Crit Rev Oncol Hematol. 2026;224:105374. doi:10.1016/j.critrevonc.2026.105374

Xu T, Dai J, Tang L, Sun L, Si L, Guo J. Systemic administration of STING agonist promotes myeloid cells maturation and antitumor immunity through regulating hematopoietic stem and progenitor cell fate. Cancer Immunol Immunother. 2023;72(11):3491-3505. doi:10.1007/s00262-023-03502-7

‍

‍

‍