Bioinformatics Seminar
Time: 3PM
Venue: Online Only
11 August 2026
This is a WEHI only event.Spatial transcriptomic analysis of primary melanomas to characterize tumour microenvironment architecture
Marie TrussartWEHI
Early-stage melanoma accounts for the largest proportion of new melanoma diagnoses. Although typically curable with surgery alone, 25% of melanoma-related deaths arise from patients with early-stage melanoma who develop recurrent disease and die as a result. There is, therefore, an urgent need to identify patients with early-stage melanoma at high risk of recurrence. However, the molecular basis underlying early-stage melanoma recurrence is not well understood. Using the 10X Genomics Visium and Visium HD spatial transcriptomics platforms, we examined 12 early-stage, thick, ulcerated (T4b) FFPE primary melanoma samples: 7 with an unexpectedly good outcome of no recurrence ≤5 years of diagnosis and 5 with an expectedly poor outcome of recurrence ≤5 years of diagnosis (late and early recurrence groups, respectively). These two groups were chosen based on the hypothesis that patients with differing and extreme clinical outcomes likely have significant differences in the underlying biology of their primary skin tumours. Spatial clustering and cell type mapping were refined using RCTD deconvolution. To characterize the immune-tumor architecture & look for differences in the underlying biology between the two groups, we performed pseudo-bulk differential gene expression, differential cellular composition using sccomp, spatial domain/niche analysis using Banksy and spatial interactions analysis using NMF. Gene set analysis revealed upregulation of the hallmark pathways associated with hypoxia, angiogenesis, epithelial–mesenchymal transition, glycolysis, IL2, TNFa and TGFb pathways in the late relative to the early group. Cellular compositions also varied across the two groups, with the late group having significantly lower tumour purity and higher abundance of immune cells than the early group. These findings suggest that distinct spatial and molecular features exist in early-stage primary melanomas, with differences in immune activation, inflammation, and metabolic programming potentially contributing to recurrence risk. Findings from this work may lead to the discovery of biomarkers predictive of recurrence, which may be integrated into clinical practice to help identify patients with early-stage melanoma at high risk of recurrence.