Nathan Wong

Background: Anaplastic thyroid cancer (ATC) is a highly lethal malignancy with poor survival and limited therapeutic options. The mechanisms driving immune evasion in ATC remain poorly understood. The cyclic GMP–AMP synthase–stimulator of interferon genes (cGAS–STING) pathway is a key mediator of innate immune sensing and antitumor immunity. Methods: We investigated the role of cGAS–STING signaling in ATC using three complementary models (i) a genetically engineered mouse model harboring triple mutations in Thrb1 , Trp53 , and Pten (RU3 model); (ii) an orthotopic tumor model derived from Trp53 - and Braf -mutant ATC cells; and (iii) subcutaneous tumor models of Ifnb1 - or Trp53 -expressing ATC cells. Tumor immune microenvironment alterations were assessed using comprehensive molecular approaches, including single-cell RNA sequencing. Results: The models demonstrated marked suppression of cGAS–STING signaling during ATC progression. This suppression was associated with impaired antitumor immunity, characterized by reduced CD8 ⁺ T cell infiltration and diminished type I interferon (IFN-I) responses. Restoration of IFNβ1 signaling in ATC cells significantly enhanced CD8 ⁺ T cell recruitment and inhibited tumor growth in vivo . These findings indicate that immune evasion in ATC is driven by decreased IFNβ1 production rather than resistance to its downstream effects. Mechanistically, p53 was found to sustain cGAS–STING–IFNβ1-mediated antitumor immunity, and its loss contributed to reduced IFNβ1 expression and immune escape. Conclusions: Impaired DNA sensing via suppression of the cGAS–STING–IFNβ1 axis promotes immune evasion in ATC. Therapeutic strategies aimed at restoring this pathway may enhance antitumor immunity and represent a promising approach for treating this aggressive disease.

Osteosarcoma is the most common malignant bone tumor in young patients and remains a significant clinical challenge, particularly at the metastatic stage. Studies detailing the immunosuppressive mechanisms within the metastatic osteosarcoma microenvironment are needed to elucidate the cellular communities in the metastatic microenvironment that support metastatic growth and to identify therapeutic approaches to target the cross-talk between cancer cells and their microenvironment. In this study, we performed spatial transcriptional profiling on a cohort of osteosarcoma pulmonary metastases from pediatric patients. The data revealed a conserved spatial gene expression pattern resembling a foreign body granuloma, characterized by peripheral inflammatory signaling, fibrocollagenous encapsulation, lymphocyte exclusion, and peritumoral macrophage accumulation. The intratumoral microenvironment of these lesions, however, lacked inflammatory signaling. Exploration of spatially distinct drug–gene interactions identified the CXCR4 signaling axis, which displayed spatial heterogeneity and complexity, as a potential therapeutic target that bridges both the intra- and extratumoral microenvironments. Collectively, this study reveals that metastatic osteosarcoma comprises multiple regionally distinct immunosuppressive microenvironments.Significance:Exploration of spatially resolved microenvironments in metastatic osteosarcoma tissues reveals how the tissue architecture promotes immunosuppression and identifies actionable processes to enhance immunotherapy efficacy.

Chemokine/cytokine antibody array profiling in metastatic osteosarcoma patient specimens and patient derived cell lines. Heatmaps showing normalized protein expression of chemokines and cytokines from (A) an independent cohort of 10 frozen metastatic osteosarcoma specimens and (B) conditioned media from five unrelated patient derived cell lines.

Spatial expression patterns of CSF1R, PTGES2, MARCO, and S100A9. Spatial feature plots showing the regional expression of CSF1R, PTGES2, MARCO, and S100A. The Osteosarcoma Signatures is shown as a reference to identify the intratumoral microenvironment. MARCO expression was either not expressed or below the quality control threshold for specimen OS-9 and thus the feature plot was omitted.

Cell type-specific marker genes

Differential gene expression of chemokine/cytokines in primary versus metastatic osteosarcoma specimens. Volcano plot of differentially expressed chemokine and cytokine genes in non-metastatic (black dots) and pulmonary metastases (gray dots) from the Sorenson, and colleagues NanoString IO360 gene expression dataset (27). Horizontal red line indicates significance threshold (q = 0.001). Dots with red outline indicates q < 0.001.