Poster Presentation 44th Lorne Genome Conference 2023

Genomic comparison of primary dermal melanoma (PDM) and subcutaneous metastatic melanoma (SCM) (#136)

Peinan Zhao 1 2 , Chris Chew 1 2 3 , Pacman Szeto 1 2 , William Berry 1 2 3 , Louise Photiou 3 , Nicholas Wong 4 , Catriona McLean 5 , Vikki Marshall 1 , Jen G. Cheung 1 2 , Victoria Mar 3 , Tony Papenfuss 6 , Mark Shackleton 1 2
  1. Central Clinical School, Monash University, Clayton, VIC, Australia
  2. Cancer Development and Treatment Group, Department of Medicine Research Laboratories, Alfred Hospital, Monash University, Melbourne, VIC, Australia
  3. Victorian Melanoma Service, Alfred Hospital, Melbourne, VIC, Australia
  4. Monash Bioinformatics Platform, Monash University, Clayton, VIC, Australia
  5. Department of Anatomical Pathology, Alfred Hospital, Melbourne, VIC, Australia
  6. Bioinformatics Division, The Walter and Eliza Hall Institute, Parkville, VIC, Australia

Background: 

Primary dermal melanoma (PDM) and subcutaneous metastatic melanoma (SCM) are typically histologically similar but prognostically distinct. Despite the impacts on patient prognosis estimation and treatment, distinguishing PDM and SCM by clinical criteria alone is challenging, and there are currently no known molecular markers that distinguish PDMs from SCMs. The primary objective of this study was to identify molecular markers that distinguish PDMs and SCMs in patients. We also aimed to infer underlying disease mechanisms based on genetic features of PDMs and SCMs.

Method:

A cohort of 20 PDMs and 23 SCMs was studied. Formalin-fixed paraffin-embedded (FFPE) blocks of PDM and SCM samples were sequenced using the Illumina TSO500 panel, which included 523 known cancer genes. Sequencing data were analysed and a predictive model of patient prognosis was established.

Results:

SCMs had significantly higher copy numbers of seven known oncogenes genes (EGFR, CDK6, FGF2, PDGFRA, BRAF, MET, MYC). Patient RFS was most closely associated (c-index=0.785, p << 0.001) with the combined features of elevated BRAF copy number and an initial clinical diagnosis of SCM. Extrachromosomal DNA (ecDNA) was identified in 3/23 SCMs patients but not in PDMs. A trend of higher copy number variation (CNV) load and lower tumour mutational burden (TMB) was found in SCM patients. 

Conclusion: 

Prediction of the prognosis of PDM and SCM patients could be improved when incorporating molecular markers into the assignment of clinical diagnosis. Compared to PDM, there were significant gains of oncogenes and cases with ecDNA in the SCM cohort, suggesting structural genomic derangement as a potential disease driver linked to metastasis.