{
  "abstract": "Background Syngeneic mouse models, with fully functional immune systems, are critical tools for evaluating anti-tumor immune responses and advancing preclinical immunotherapy strategies. However, the immune landscape differs markedly between models, making the selection of appropriate systems for specific therapeutic investigations complex. To better understand tumor-immune interactions in this context, we applied multiplexed spatial proteomics to untreated syngeneic mouse glioma, GL261 and lung carcinoma, LL/2 models.Methods Using the Cell DIVE™ Multiplexed Imaging Solution and a novel panel of IHC-validated, mouse-reactive antibody conjugates from Cell Signaling Technology® (CST®), we spatially profiled immune and tumor markers within FFPE (Formalin-Fixed Paraffin-Embedded) tissue slides from GL26 and LL/2 syngeneic mouse tumors. Tisue sections were iteratively stained with the validated CST® antibodies, and in every round, the tissue was imaged on the Cell DIVE Imager using four channels plus DAPI, with automatic AF removal, corrections, and stitching. Fully stitched imaging data were analyzed using Aivia 15, an AI-powered image analysis software, for precise segmentation, phenotyping, and clustering of immune cell populations.Results Overall, Cell DIVE multiplexed imaging using CST® antibodies, and AI-guided image analysis using Aivia revealed distinct immune signatures and spatial patterns unique to each tumor model, highlighting the heterogeneity of immune infiltration and tumor-immune dynamics. First, AI-powered segmentation and phenotyping on Aivia revealed alterations in the abundance of diverse immune cell types and immunosuppressive populations in GL26 and LL/2 syngeneic tumor tissues. Next, clustering analysis showed differences in co-expression profiles of key phenotypic markers and functional markers. Finally, neighborhood analysis of tumor-immune populations determined key differences in the positioning of various immune cell types in GL26 and LL/2 tissues. Together, this data provides insights into how differences in immune cell phenotype, function, and organization across tumor models may influence the anti-tumor immune response.Conclusions This approach underscores the value of combining multiplexed spatial imaging with AI-driven analysis to enhance model selection, interpret therapeutic response, and identify spatially resolved biomarkers. Our findings support the application of advanced spatial proteomics and computational tools to deepen understanding of the immune landscape in solid tumors and accelerate development of immune-based therapies.",
  "authors": [
    {
      "affiliations": [
        "Leica Microsystems, Waltham, MA, USA"
      ],
      "name": "Vasundhara Agrawal"
    },
    {
      "affiliations": [
        "Cell Signaling Technology, Danvers, MA, USA"
      ],
      "name": "Emily Quann Alonzo"
    },
    {
      "affiliations": [
        "Leica Microsystems, Waltham, MA, USA"
      ],
      "name": "Sophie Struble"
    },
    {
      "affiliations": [
        "Cell Signaling Technology, Danvers, MA, USA"
      ],
      "name": "Ryan Sinapius"
    },
    {
      "affiliations": [
        "Cell Signaling Technology, Danvers, MA, USA"
      ],
      "name": "Gabriella Spang"
    },
    {
      "affiliations": [
        "Cell Signaling Technology, Danvers, MA, USA"
      ],
      "name": "Shelby Sponaugle"
    },
    {
      "affiliations": [
        "Cell Signaling Technology, Danvers, MA, USA"
      ],
      "name": "Jeremy Fisher"
    },
    {
      "affiliations": [
        "Leica Microsystems, Waltham, MA, USA"
      ],
      "name": "Richard A Heil-Chapdelaine"
    },
    {
      "affiliations": [
        "Leica Microsystems, Waltham, MA, USA"
      ],
      "name": "Natasha Fernandez Diaz Granados"
    },
    {
      "affiliations": [
        "Leica Microsystems, Waltham, MA, USA"
      ],
      "name": "Arindam Bose"
    }
  ],
  "title": "62 Multiplexed spatial proteomics with AI-guided analysis reveals tumor-immune architecture in syngeneic mouse glioma (GL261) and lung carcinoma (LL/2) models",
  "uid": "1ebc4e3a-48e8-50ff-80d3-777b9441bbf3"
}