Epithelial-mesenchymal transition and immune exclusion underpin the natural history of primary cutaneous melanoma
Need to claim your poster? Find the KiKo table at the conference and they'll help
you get set up.
Presented at: Society for Investigative Dermatology 2025
Date: 2025-05-07 00:00:00
Views: 2
Summary: Abstract Body: Although outcomes in primary melanoma are chiefly dictated by Breslow depth, the pathways through which thicker tumors predispose to metastasis and recurrence are poorly described. This translational study leveraged complementary omics modalities (Visium spatial RNAseq, Chromium single-cell RNAseq, and CODEX spatial protein profiling) on multiple primary melanoma cohorts to characterize the natural history of thin versus thick melanoma. We first generated a single-cell dataset for primary cutaneous melanoma from Cohort A (Chromium, n = 50) - currently the largest single-cell reference for primary melanoma in the literature. This was employed for cell type deconvolution in our transcriptomics cohort. On pseudobulked GSVA of Cohort B (Visium, n = 101), thick primary tumors (>1mm Breslow depth) demonstrated enrichment of EMT- (BH-adj p: 0.019) and E2F-associated gene sets(adj p: 0.008) in melanoma spots as compared to thin tumors. Thick tumors exhibited an immune-excluded phenotype relative to thin tumors, with a lower ratio of intra-tumoral:peri-tumoral lymphocytes (adj p: 0.017) and a higher median NN distance between tumor cell-T cell pairs (adj p: 0.037). Ligand-receptor analyses indicated greater spatial interaction between immunosuppressive gene pairs in thick versus thin melanomas (CD163-TWEAK, adj p: 0.001; CD160-HVEM, adj p: 0.042). These features persisted in T1b vs T1a melanoma in Cohorts C (Visium, n = 54) and D (CODEX, n = 6), suggesting that phenotypic delineation by tumor thickness is present from the earliest stages of the disease. On Cox regression. EMT enrichment (aHR: 2.83; p: 0.007), E2F enrichment (aHR: 2.19; p: 0.032), and T cell exclusion (aHR: 2.56; p: 0.025) were each correlated with lower recurrence-free survival in Cohort B (min. 7y follow-up). These spatially derived metrics capture biologically relevant tumor features that may partially mediate the association between tumor stage and recurrence in primary melanoma. Samuel X. Tan<sup>1</sup>, Nicholas M Muller<sup>1</sup>, Chenhao Zhou<sup>1</sup>, Nisal Vipulaguna<sup>1</sup>, Yung-Ching Kao<sup>1</sup>, David C. Whiteman<sup>2</sup>, H P. Soyer<sup>1</sup>, Mitchell S. Stark<sup>1</sup>, Quan Nguyen<sup>2</sup>, Kiarash Khosrotehrani<sup>1</sup> 1. The University of Queensland, Brisbane, QLD, Australia. 2. QIMR Berghofer, Australia, QLD, Australia. Translational Studies: Cell and Molecular Biology