876 Fibroblast-immune cell interactions conserved between the healing wound and tumors: their identification and implications for the anti-tumor immune response

<h3>Background</h3> The idea that tumors co-opt elements of tissue repair and maintenance for growth and immune evasion is a long-held one.¹ Cancer associated fibroblasts or CAFs can control the localization and polarization state of infiltrating immune cells, therefore they represent an enticing target for clinical modulation.^2–4 Through mechanisms including ECM modulation, cytokine secretion and even antigen presentation⁵, these relatively rare but powerful cells have the potential to affect the immune response against tumors. Which fibroblast states promote or inhibit the anti-tumor response, how do these states arise and where are these cells localized relative to specific immune cell subpopulations? <h3>Methods</h3> We tackled these questions through an integrative approach combining scRNA-Seq, imaging and spatial transcriptomics on the healing wound and tumor models.⁶ Our approach focused on identification of gene programs, suites of genes with co-expression, rather than cell subclusters to pinpoint cell-cell interactions in our datasets. Spatial and temporal context to our scRNA-Seq datasets further added confidence to our cell-cell interaction predictions. <h3>Results</h3> We began with generation of a spatially and temporally resolved single cell RNA-Seq dataset of both immune and non-immune cell subsets of a healing skin wound. Using this dataset, we identified shared space-time patterns of cell behavior encompassing multiple cell types at once. Specifically, we verified through large-volume imaging and in-vitro studies several communication axes between fibroblasts and macrophages including OSM, TNC and POSTN mediated interactions. Analyses of mouse tumor models and patient data revealed these interactions were likely conserved during tumor progression. Further examination of spatially resolved mouse melanoma datasets^{7 8} allowed us to zoom in on a cellular network enriched at the edge of the tumor composed of fibroblasts and edge-restricted myeloid populations. We find that disruption of this network correlates with poorer CD8⁺ T cell infiltration into the tumor and a more exhausted T cell compartment. To further dissect the formation and importance of this network, we applied in vivo genetic perturbations and multiplexed imaging. <h3>Conclusions</h3> We hypothesized that tissues undergoing pathological perturbations may exhibit fundamental, conserved cell-cell interactions and linking these distinct tissue contexts can yield useful inferences. We demonstrate that fibroblast-immune crosstalk mechanisms can be found in both healing wounds and tumors. This conservation led us to investigate fibroblast-immune interactions in the tumor and single out a unique network of fibroblast-myeloid cells that alter CD8⁺ T cell infiltration into tumors. <h3>References</h3> Dvorak Harold F. Tumors: wounds that do not heal. <i>New England Journal of Medicine</i> 1986;<b>315</b>:1650–1659. Krishnamurty AT, <i>et al</i>. LRRC15+ myofibroblasts dictate the stromal setpoint to suppress tumour immunity. <i>Nature</i> 2022;<b>611</b>:148–154. Barone F, <i>et al</i>. Stromal fibroblasts in tertiary lymphoid structures: a novel target in chronic inflammation. <i>Front Immunol</i> 2016;<b>7</b>. Barrett R, Puré E. Cancer-associated fibroblasts: key determinants of tumor immunity and immunotherapy. <i>Current Opinion in Immunology</i> 2020;<b>64</b>:80–87. Kerdidani D, <i>et al</i>. Lung tumor MHCII immunity depends on in situ antigen presentation by fibroblasts. <i>J Exp Med</i> 2022;<b>219</b>:e20210815. Hu KH, <i>et al</i>. Transcriptional space-time mapping identifies concerted immune and stromal cell patterns and gene programs in wound healing and cancer.<i> Cell Stem Cell</i> 2023;<b>30</b>:885–903.e10. Hu KH, <i>et al</i>. ZipSeq: barcoding for real-time mapping of single cell transcriptomes. <i>Nat Methods</i> 2020;<b>17</b>:833–843. Ray A, <i>et al</i>. Critical role of CD206+ macrophages in organizing anti-tumor immunity. 2023.10.31.560822 Preprint at https://doi.org/10.1101/2023.10.31.560822 (2023).