Exploring novel approaches to metastatic heterogeneity and clonal evolution

Abstract

Background Alternative methods are required to optimise the clinical insights that may be derived from studies of cancer evolution and heterogeneity. Profiling a single biopsy of tumour tissue underestimates potential genomic diversity. In melanoma, studies in this field have been limited by poor access to primary tumours, small cohorts and a bias to locoregional metastases. Through the use of extensive sampling at post-mortem and an analysis of homogenization of 'leftover' surgically resected tumour, I aimed to explore the potential unity of two novel strategies. Research plan An in-depth exploration of modes of metastatic seeding as well as inter-metastatic heterogeneity was facilitated by sampling multiple sites from a cohort of advanced melanoma patients undergoing a research post-mortem, coupled with analysis of primary tumours where possible. I also designed and implemented a pan-cancer study where non-embedded, leftover surgically resected tumours were 'homogenised' (blended) with a view to determine the feasibility of this approach as a potential tool to inform treatment decisions. Conclusion Polyclonal seeding dominates primary to metastatic evolution in melanoma. This leads to greater inter-metastatic heterogeneity than previously appreciated, which may underlie the clinical behaviour of certain sites of disease in response to treatment. Genomic mechanisms of treatment resistance may be heterogeneous and are insufficient to explain the majority of progression events on targeted and immune checkpoint treatment. Homogenisation of leftover tumour samples appears to be feasible approach in primary, breast, renal and tubo-ovarian cancers, as well as sarcomas and melanoma lymph node dissections. Through its inclusion of a greater volume of tissue, it is likely to yield greater depth of genomic information than routine molecular profiling of small samples, although further work is required to confirm this.