Spatial transcriptomics reveals epithelial-immune remodeling preceding malignant transformation of oral premalignant lesions

Oral premalignant lesions (OPLs) pose challenges to histopathological grading as they fail to accurately predict cancer development. To uncover the tissue ecosystems that lead to malignant transformation, researchers combined single-cell-resolution Xenium spatial transcriptomics from 20 HPV-negative biospecimens sourced from 16 patients with OPLs, alongside independent single-cell RNA sequencing and immunofluorescence data from 38 patients experiencing OPLs.

The findings reveal a coordinated epithelial-immune program that sets the stage for cancer progression.

This progression involves distinct basal epithelial states characterized by high levels of MX1 and NOTCH3, alongside a surge in S100A9-high inflammatory macrophages. The landscape also features TIGIT-high, exhaustion-associated T cells and altered dendritic cell states. Crucially, spatial analyses indicate that T cells and dendritic cells have been displaced from their original position at the basal epithelial interface.

Furthermore, epithelial neighborhoods display a depletion of dendritic cells but show an enrichment for TIGIT-high T cells, suggesting an immune reorganization that is not solely dependent on cellular abundance.

Separate analyses using single-cell data and protein-level information corroborated the basal epithelial interferon/stress programs, S100A9-associated myeloid inflammation, TIGIT-high, exhaustion-associated T-cell programs, and the redistribution of dendritic cells. By inferring ligand-receptor interactions, researchers identified convergent myeloid and lymphoid signals linked to epithelial stress, checkpoint regulation, and extracellular matrix remodeling.

Functionally, recombinant S100A9 was found to accelerate wound closure in both oral epithelial and cancer cells. Conversely, S100A9 and S100A8/S100A9 enhanced oral cancer cell proliferation, responses that were mitigated by pharmacologic inhibition of TLR4 or RAGE. These results collectively paint a picture of a spatially organized, myeloid-skewed, checkpoint-enriched ecosystem that exists prior to cancer invasion.

The study suggests that S100A9-TLR4/RAGE signaling and alterations in epithelial-immune geometry could serve as potential candidate mechanisms and biomarkers for early detection and interception of oral cancer.